by DR. GEORGE ONODA

Faster ThanA Speeding Bullet?

How fast does the cue ball really travel on a typical break shot?

"HE'S GOING TO knock this ball in at100 miles per h o u r . . . , " whispered anESPN announcer during a televised 9-ballmatch.

No, the announcer wasn't Joe Isuzu.But, the inference that the cue ball cantravel 100 MPH made me curious. Justhow fast does a professional pool playerpropel the cue ball on a 9-ball breakshot?

I asked some of my pool playingfriends, and their guesses ranged fromspeeds of 50 to 150 miles per hour. SinceI had never seen this actual speedreported, I set off lo make suchmeasurements.

Video tapes are a rich resource of topprofessional players in action. Withmodern technology, it is now possible toanalyze these tapes in great detail, usingslow motion and frame-by-frame stop ac-tion. The time between each frame is one-thirtieth seconds.

The average speed of a ball betweentwo points can be measured from a videotape by knowing the distance traveled andmeasuring the time taken to travel thisdistance. Since the dimensions of a tableare fairly standardized, the distance bet-ween the starting cue ball position and theloot spot where the one ball lies can beeasily determined. For example, if the cueball is placed on the head string, one dia-mond to the right of the head spot, thedistance to the foot spot is typically 53inches.

The time of travel is obtained from thenumber of video frames taken frombeginning to end. The position of the cueball in each frame is marked as a dotdirectly on the video screen. The result isa series of dots along the path. The spac-ing of the dots in the middle of (he pathare measured. From this we can deter-mine the total number of frames from thebeginning to the end. The time from startto finish is equal to the number of framestimes one-thirtieth second. Dividing thisinto the total distance gives the averagespeed of the ball in inches per second.This is converted to miles per hour. Themeasurements are accurate to around plusor minus one-quarter miles per hour.

To be certain that the frame countingmethod gives the right answer, spotchecks were made using an entirely dif-ferent technique. Here I used the sounds

made when the cue hits the cue ball andwhen the cue ball hits the rack. With aspecial oscilloscope, which plots thesound intensity versus time, the time oftravel (from one sound impulse to thenext) could be accurately measured. Itwas found that this sound method and theframe counting method gave the sameresults.

These methods measure the averagespeed of the cue ball. Because the cue ballactually slows down gradually, the initialspeed is slightly higher and the final speedis slightly lower than the average speed.But since the speed only diminishes by afew percent, the average speed serves ade-quately as a measure of the cue ball speedin a break shot.

1 analyzed video tapes of some of thematches in the 1986 Resorts Internationaltournament, and the 1987 Brunswick9-Ball Team Challenge. (Players involv-ed in the tapes included Mike Sigel, JimRempe, Allen Hopkins, Steve Mizerak,Nick Varner and Jose Garcia.) The speedof break shots ranged from 22 to 26 MPH,the average being 24 MPH. The top speedof 26 MPH was recorded for several breakshots by Mike Sigel. At this speed, the cueball reached the one ball in 0.12 seconds.

The speeds of break shots in severalwomen professionals were also measured,using a tape of the Brunswick TeamChallenge. In one break shot Jean

Balukas propelled the ball at 22 MPH.Ewa Mataya, Belinda Bearden and LoreeJon Jones typically had speeds between18 and 21 MPH. From this small sampl-ing, the women's speed were on theaverage around 20 MPH, which is about83 percent of the average for the men.

The three or four MPH variations inbreak shot speeds within players of thesame sex might account in part for whysome players have better break shots thanothers. But the tapes reveal that success,where a ball is made, was not stronglycorrelated with speed. Accuracy wasequally important, which meant hittingthe one ball in the right place and withthe proper English. Often in successfulbreak shots, the cue ball was observed tomove about less after hitting the rack thanwith the unsuccessful shots, indicatingthat more of its energy has been transfer-red to the object balls.

In any event, the surprising truth is nowrevealed. The top speed that a profes-sional propels a cue ball is around 26miles per hour. Sorry to say that it is not100 miles per hour, which is the top speedfor throwing a baseball. While this maydamage egos, take comfort in the factthat the cue ball weighs twice that of abaseball.Dr. George Onoda is a research scientist alIBM's Research Division, as well as an avidpool and billiard player.

The truth is finally out: Break shots by pros like Danny DiLiberto (above), travel around 25 MPH.

34

Perfect Draw?Test your draw shot against the best.

by DR. GEORGE Y. ONODA

STROKING THE BALL as low aspossible is like sliding a coin to theedge of a table. The best and the worstdiffer by a gnat's eyelash. In pool, youeither get a great draw or you miscue.Few players chance cutting it too close.But the closer you approach the brink,the better is the quality of your drawshot.

How low do you strike the cue ball?This cannot be determined from theusual draw test, where you measurehow far a ball is drawn back in a head-on shot with an object ball. The drawdistance depends not only on how lowyou hit the ball, but also on the forceyou apply and on the speed of thecloth. To reveal how low the cue ballis being struck, I have developeda method that you can easilycarry out on a pool table.

The setup for the method isillustrated in Fig. 1 (a regulationpool table is assumed). An ob-ject ball is placed 4% inches fromthe foot spot in the direction ofthe foot rail. Applying draw tothe cue ball, the object ball isshot into the upper-right cornerpocket. The cue ball is placedwhere the aiming direction isacross the table, parallel to thefoot rail. The cue ball must beas close as possible to the objectball, without interfering with thestroke, so that the least amountof back spin is lost when the ob-ject ball is reached.

When the cue ball caroms offthe object ball (Fig. 1), it initial-ly travels along the tangent line(a line perpendicular to thedirection that the object balltravels). Then the back spincurves the ball to the left, towardthe head rail. After curving, theball rolls along a straight path,or track.

The direction of this finaltrack determines the quality ofthe draw. The closer this direc-tion is to being parallel to thelong rail, the lower the cue ballhas been hit below center dur-ing the stroke. Altering the speedof the shot causes the length ofthe curve to change and the finaltrack to shift. However, if thecue ball is struck the samedistance below center, the direc-

tion of the final track remains the sameirrespective of speed (see Fig. 2). Thespeed of the cloth also shifts the trackbut does not change its direction. Thismethod works well with soft andmedium strokes. If the ball is hit toohard, it will strike the long rail whileit is still curving, and the final direc-tion is not achieved.

This method is based on theory (thephysics of pool). Let us suppose forsake of argument that the lowest a cueball can be struck safely is around two-fifths the distance from the ball'scenter to the table surface. For the pre-sent, we will call this the perfect draw.Theory predicts that when the objectball is cut 39 degrees from the forward

direction, the final track of the cue ballstroked with perfect draw and ahorizontal cue will be perpendicular tothe forward direction. In our setup, theobject ball goes off at 39 degrees ifsunk in the corner pocket. Since thecue ball is aimed parallel to the shortrail, its final track ends up parallel tothe long rail when a perfect draw shotis achieved.

Thanks to friends at the Jack andJill Cue Lounge (Brewster, N.Y.), thismethod has been tested with players ofvarying abilities. The best players couldsometimes, but not always, bring theball parallel to the long direction. Less-skilled players had inclined trackswhere the cue ball hit the long rail

1 4 Billiards Digest • May/June 1989

before reaching the head rail. In-terestingly, one player with an excep-tional stroke seemed to be able to ex-ceed the parallel limit by a few degrees.This might mean that the ball wasstruck lower than what we assumed forperfect draw. However, there is a littleslop in our test, because the width ofthe pocket is more than twice thediameter of the ball. It is possible tocheat the pocket by about two degrees,which could explain how the parallellimit was exceeded slightly.

With this method, it is possible totest the truth about several assertions.One is that the cue ball can be hitsignificantly below the two-fifthspoint. If the ball could be struck halfthe distance from the center to the bot-tom, then the final track would exceedthe "parallel limit" by around 12degrees. This would be a shift sidewaysof almost one diamond for a forwardtravel of four diamonds, which wouldbe very noticeable. There is also thequestion as to whether a partially in-clined cue aids in increasing draw. Inour setup, the cue must be inclined(slightly elevated) by at least fivedegrees because of the interference ofthe rail. If this causes an increase inthe amount of draw compared with ahorizontal cue, then increasing it evenmore (e.g. 20 degrees) should have apronounced effect and cause the trackcourse to exceed the parallel limitsignificantly. (A masse shot with anear-vertical cue is another matter,which we do not include in thisdiscussion.)

So the question is, can the parallellimit of the track be exceededsignificantly by any means other thana strong masse shot? Is it possible tocause the track to point toward thebottom-left corner in Fig. 1, for exam-ple? If any reader can achieve thisseemingly superhuman draw, let meknow, as it will give us new insights onthe physics of draw. Please be remind-ed that the test is valid only if: 1. Theballs are in the right position; 2. Youraim is parallel to the short rail; and 3.The object ball is sunk in the cornerpocket. It is helpful to have someonewatch to be certain that you are aim-ing parallel to the short rail as youshoot.

Dr. George Onoda is a research scientist atIBM's Research Division, as well as an avidpool and billiard player.

Billiards Digest • May/June 1989 1 5

by Dr. George Onoda

Know TheseWinning Caroms

Win with a dead carom off a spotted object ball.

CONSIDER THE SITUATION inDiagram 1 from a 9-ball game. Youropponent made the 5 ball, but hasscratched. The 5 is on the foot spotand you have ball-in-hand. The 9 ballis sitting in the jaws of the far, cornerpocket. You cannot shoot the 5 intothe 9. Nor can you make the 5 ball ina pocket to start a run. Your best hopeis either to play a safety or to caromthe cue ball off the 5 and into the 9for an immediate win.

Many players will pass up the caromshot because the opponent may runthe table if it is missed. It is easy toplay an effective safety instead, givingyou a chance to run the table or knockthe 9 in on your next turn. But youmight not be so lucky; your opponentmay play a great kick shot, leaving youwith nothing.

You probably would not hesitateplaying the carom immediately if itwere a high-percentage shot. Howwould you play the carom? Wouldyou put the cue ball at point N and hitthe 5 ball very thin? This is a delicateshot, with little margin for error inaiming the ball; you would be doingwell making this shot 50 percent of thetime. Would you instead put the cueball near point P and carom with drawor stun action? This is an even moredifficult carom to execute reliably, forthe path depends on three variables(the angle of the cue, the amount ofback spin, and the speed).

So, if the caroms from N and P arenot reliable enough for this situation,is there a better carom shot? Definite-ly! In fact, there is a carom that youcan make better than 90 percent of thetime. It involves a half-ball hit on the5 ball, with soft, rolling action, bothof which are easy to execute. But youhave to know how to set it up.

First, why a rolling, half-ball hit? Asdiscussed in several of Robert Byrne'sarticles, a soft, rolling, half-ball shotgives a deflection angle for the cue ballthat is accurate, reliable and resistantto error. This deflection angle is now

accepted to be 34 degrees. It is alsoeasy to roll a cue ball; you only needto stroke it softly above center. Andthe aim does not have to be perfect,since a small error does not change thedeflection angle significantly.

The above knowledge is of littlevalue unless the cue ball and two ob-ject balls are precisely aligned for the34-degree deflection angle. In mostgame situations, you have little or no

control over the alignment. But if youhave cue ball in hand, you can set upwhatever alignment you wish (assum-ing no interfering balls). In otherwords, in Diagram 1, there is a place(e.g. Point R) where the cue ball willbe properly aligned with the 5 and 9balls for the easy carom. The only pro-blem is knowing where to place the cueball.

Diagrams 2, 3 and 4 show how to

20 Billiards Digest • December 1989

set up the proper alignment for caromsleading to different pockets on thetable. These are for pockets on oneside of the table. For those on theother side, imagine looking at thediagrams in a mirror. Now eachpocket in the three diagrams can bereached in two ways, from one side orthe other of the object ball. Criticalpoints along the rails are identified.Once an appropriate critical point isknown, a critical line is drawn fromit to the edge of the object ball. Thecue ball is then placed anywhere alongthis critical line, the spotted ball isstruck with a soft, rolling, half-ball hiton the appropriate side of the objectball, and the cue ball follows thecarom line to the 9 ball.

To reach a near corner pocket, asshown in Diagram 2, one critical

point, "A," lies on the opposite longrail, 1 inch past the second diamond.The other critical point, "B," is on thehead rail, one-half diamond from thecenter diamond. This shot from PointB is essentially the shot de-scribed on page 112 of Byrne's Stan-dard Book of Pool & Billiards. Whenthis book was originally published, thestandard 9-ball scratch rule was thatthe cue ball was in hand behind thehead string. Now, with the morepopular ball-in-hand anywhere on thetable rule, dead caroms to all pocketsbecome possible.

To reach a side pocket, as inDiagram 3, one critical point, "C ,"lies 1 inch from the center diamond onthe foot rail, in a direction away fromthe side pocket. The other point, "D,"lies 1 inch from the second diamond

on the opposite long rail, in the direc-tion toward the foot rail.

To reach a far corner pocket,Diagram 4, one critical point, "E,"lies 1 inch from the first diamond onthe foot rail. The second point, "F ,"lies on the opposite long rail, halfwayfrom the first diamond and the cor-ner pocket.

These critical points are not hard toremember, since they are either 1 inchfrom a diamond, or are halfway bet-ween two diamonds. Their locationsshould be checked for your table, asvariations can occur with differenttables. The necessary adjustments canbe determined quickly by trial anderror.

I know that some readers maydoubt the reliability of these shots. Butif you are careful in aligning the cueball along the critical line, if you shoota fairly accurate, half-ball hit, and ifyou shoot softly, it's almost impossi-ble to miss. You will be a believer ifyou try it. With a few minutes of prac-tice, you can make these shots at leastnine out of 10 times.

The carom is somewhat more dif-ficult if the 9 ball is not squarely linedup in the pocket. A compensation ismade by moving the cue ball off thecritical line a small amount. Tovisualize what is necessary, imaginethat the critical line and the carom lineare joined together as a stick, bent 34degrees at the foot spot and nailed tothe foot spot about which it couldrotate. If we move the near end of thestick to the left, the far end moves tothe right, and vice versa. Thus, if the9 ball is misaligned to the left of thecarom line, the cue ball must be mov-ed to the right of the critical line. Theamount moved depends on how farthe 9 ball and the cue ball are from thefoot spot. The further the 9 ball, theless the cue ball has to be shifted.Judging these shifts requires practice,but can be quickly learned.

Opportunities such as that depictedin Diagram 1 do not arise often. Butneither do dead-on, frozen combina-tions for the 9 ball. Nevertheless, ifyou recognize and know how to takeadvantage of such gifts, you will winmore games. A game-winning carommight even break your opponent'sconcentration.Dr. George Onoda is a research scientist atIBM's Research Division, as well as an avidpool and billiard player.

22 Billiards Digest • December 1989

by Dr. George Onoda

Analyzing The 9-Ball BreakUseful insights on where balls travel.

THE TYPICAL 9-BALL break shotis a slam-bang explosion that sendsballs rushing in every conceivabledirection at break-neck speeds. Norm-ally, players and spectators alikehaven't a clue as to the carnage untilthe balls have stopped rolling. But arethere important trends in the path ofvarious balls that we can use to ouradvantage?

To seek some answers, I began adetailed, systematic analysis of tour-nament video tapes, using slow motionand stop action.

Before getting into the findings ofthis study, first a few words aboutnotations, definitions and assump-tions. Ball positions in a rack will bedenoted by the letters "A" through" I " (Diagram 1). "Collisions" willrefer to contact between two objectballs after the initial rapid impacts oc-curring within the rack, "Directly"pocketed balls will be those reachingthe pocket without collisions and willinclude those made by bank shots. The"top" of the table is the foot end,while the "bottom" of the table is thehead end. Pockets are identified bytheir positions ("left-top," "left-bottom," etc.) The forward directionis toward the top, the backward direc-tion is toward the bottom. The cue ballis assumed to be shot from the rightside of the table and directly (full) atthe 1 ball.

The relative speeds and directions ofthe object balls leaving the rack areawere observed in hundreds of breakshots. Interestingly, a number ofsystematic trends were seen, as sum-marized in Diagram 2. The fans in thediagram, emanating from the centersof each object ball, designate relativespeeds and directions. The longer thefan, the greater is the relative speed.The spread of the fan indicates therange of directions that were observed.

As with player introductions, let'smeet the balls in a 9-bali rack:

Ball "A" (always the 1 ball) movesto the left and back with slow tomoderate speed. Its exact pathdepends on where the I ball is hit bythe cue ball. If hit toward the center,

the 1 ball moves in the direction of theleft-bottom pocket. If hit more on theright side, the 1 ball moves moretoward the left-side pocket. The 1 ballcan be directly pocketed in one ofthree pockets. It can fall in the left-sideor left-bottom pocket, or be bankedinto the right-bottom pocket. If notpocketed or diverted by a collision, the1 ball usually ends up at the bottomhalf of the table.

Ball " B " travels along a narrow

path, to the left and somewhat back.In some instances, it banks into theright-side pocket. Otherwise, it tendsto remain in the middle part of thetable, unless kicked away.

Ball "C" travels weakly to the right,either sideways or forward. Because ofits slow speed, it does not reach thetop-corner pocket and tends to remainin the top-right part of the table,unless kicked away.

Ball " D " is the fastest ball leavingthe rack. It was never observed totravel directly into the top-left pocket;rather, it strikes the long rail, withinone diamond of the pocket.

Ball "E" (the 9-ball) is probably themost unpredictable ball. It movesslowly in a generally forward direc-tion, to the left and sometimes to the

right. Occasionally it manages to reachthe left-corner pocket; rarely does itreach the right-corner pocket. Whenthe 9-ball is made on the break, it isoften kicked in by another ball.

Ball 'F" is the second fastest ball toleave the rack, traveling toward theright-top corner. It either strikes thelong rail or falls directly into the top-right pocket.

Ball "G" moves weakly forwardand to the left. If not hit by anotherball, it often remains near the top railand the left side. This ball is notpocketed unless kicked by anotherball.

Ball "H" travels forward and to theright with moderate speed. Ifunobstructed, it banks off the top railand moves toward the right-centerarea of the table. It can occasionallybe sunk in the right-side pocket.

Ball " 1 , " the third fastest ball leav-ing the rack, banks off the top rail.The further the 1 ball is truck on theright side, the more " I " travels to theright. If the " I " ball escapes collisions,it moves to the right-bottom end of thetable, where it could fall in the cornerpocket.

Most top players break from theside, within one diamond of the longrail. The does not increase the chancesof making ball "F , " the most fre-quently pocketed ball. Neither is thereconclusive evidence that it increasesthe chances of making the 9 ball.However, it most definitely increasesthe chances of making the 1 ball in theleft-side pocket (the second most fre-quently pocketed shot). The 1 ballmoves toward the side pocket if it isstruck at the appropriate point on theright side of the cue ball. This is theonly ball where a player has a signifi-cant opportunity to influence the out-come. Also, breaking from the sidecauses ball " I " to bank toward theright, increasing its chances of fallingin the bottom-right corner.

The majority of shots made on thebreak are due to directly pocketedballs, while collisions account for therest. As expected, the fast-movingballs (D, F and I) account for most of

26 Billiards Digest • February 1990

the lucky collisions. Balls D and Fbank around the top corners and col-lide with slower balls, such as G andH. Ball I banks off the top rail andkicks balls such as H and C. As theresults of these collisions depend onminute differences in timing and path,little prediction is possible.

A widely-held strategy is to try andleave the cue ball in the center of thetable after the break because this in-creases the chances of having a shotat the 1 ball. Since the analysis showsthat the 1 ball usually ends upsomewhere in the bottom half of thetable, the soundness of this logic issupported.

How can this knowledge affect agame?

Well, when you rack balls, do youpay any attention to how you arrangethem? Several strategies have beenbrought to my attention by colleaguesRobert Byrne, George Cleveland andSteve Priest, among others. Supposeyou want to increase the chances thatthe 2 ball will lie near the top rail afterthe break, so as to make the transi-tion from the 1 ball to the 2 ball moredifficult for your opponent. Theanalysis indicates that position G ismost likely to produce this result. Ifyou also want the 3 ball to travel tothe bottom end of the table, positionsI or B seem favorable.

The placement of balls could alsoaffect a handicap game. It you don'twant your ball to be sunk by your op-ponent, positions C and G appearsafest. If your opponent changes theside from which he breaks, these posi-tions become B and H, both stillrelatively safe. In situations where it

is favorable to you if your opponentsinks a certain ball, place it at F.

Most players have their ownstrategies in breaking balls. Hopeful-ly, the information I have providedwill help assess these strategies. Iapologize to those readers hoping I

would reveal how to make the 9 ballon the break.

Even if I knew that secret, would Igive it out for free?Dr. George Onoda is a research scientist atIBM's Research Division, as well as an avidpool and billiard player.

28 Billiards Digest • February 1990

by Dr. George Onoda

Seeing The PointOf A Frozen Kiss Shot

A sighting method for visualizing the object ball's path.

IN MOST WALKS of life, a cold kissis a truly underwhelming experience.

But pool is very special, and fewaspects of the game are as delicious tobehold as a "cold" kiss shot ... at least,to the accomplished player. Let's seewhat you can do to improve your abili-ty to determine whether kiss shots arereally "on" or "off."

Diagram 1 shows frozen balls A andB. Line A-B is drawn through theircenters. If ball A is struck by a cue ballas shown, its path of travel will be 90degrees (or perpendicular) to line A-B. (This relates closely to the conceptdiscussed recently by almost allBilliards Digest instructional writers:on all cut shots, a cue ball struck withstop shot action will deflect 90 degreesfrom the path of the object ball.)

In any case, it's obvious that if youexpect to pocket ball A, or any otherkiss shot, you'll have to visualize that90° direction accurately. Some playersappear to do this with no conscious ef-fort at all. One conscious technique isto point at the A-B line with your cue,and extend your hand across it (asthough making an oversize plus sign).If your hand points to a pocket, you'rein business. Another is to stand at thepocket you want for the shot, andvisualize an imaginary line (or placeyour cue along it) from the center ofthe pocket to the contact point be-tween balls A and B. If that line andthe A-B line seem perpendicular, or ifthe two balls seem to be the samedistance from the pocket, that kissshot is "on."

If you find these visualizations dif-ficult or unreliable, try this "point"method: bend down and sight betweenballs A and B in the general directionof the pocket (Diagram 2). The "V"between the balls acts as a rifle sight.Move your eyes to the left and right,looking at the contact point. Whenthat point looks sharp to you(Diagram 2A), you're sighting alongthe 90-degree line. If your line of sightis off, the two balls will appear tooverlap (Diagram 2B). It should be ob-vious that you do nothing aboutshooting ball A until you're correctlysighted.

Notice, in Diagram 1, that it is theedge of ball A - not the center - whichfollows the 90-degree sight line. Thepath of the center of ball A is shiftedby half the width of the ball. So your

sighted line should correctly point toa pocket's edge, not its center.

Judging kiss shots is a matter ofgood lighting, fairly keen eyesight, andpractice. Experiment with the techni-ques described here and see whichworks best for you; one of them isalmost certain to.

One important word of caution: inDiagram 1, note the direction of thecue ball. In order to move ball A 90degrees from the A-B line, the cue ballmust make contact well off to the sideof the A-B line. If ball A is struck toofull, relative to the A-B line, you'll getmarkedly different results. The pathof Ball A deviates forward from the90-degree line. Just why and how willbe discussed in a future article.Dr. George Onoda is a research scientist atIBM, and an avid pool and billiard player.

28 Billiards Digest • April 1990

by Dr. George Onoda

Play For Show,Or Play For Dough

Winning 9-ball games with safety play.IN MOST 9-BALL tournaments, youhave two ways to win a game: you sinkthe 9-ball legally, or your opponent"scratches-out" (that is, scratchesthree times in a row). In professionalplay, almost all games are won bymaking the 9-ball. Safeties are used forvarious tactical reasons, but they areseldom used to force a scratch-out. Inamateur play, on the other hand, thereare more opportunities to win gamesby forcing a scratch-out. Here's whythis is so, along with some examplesof winning with safety plays inamateur competition.

Forcing a scratch-out on a profes-sional is far more difficult than on anamateur. A pro seldom scratches onhis own, while an amateur does so oc-casionally and sometimes frequently.A pro can usually "kick" successful-ly at a ball; amateurs have less successat that. It is especially unlikely that apro would miss a kick two or threetimes in a row, while many amateursare quite capable of such misplays.

So if you compete at the amateurlevel, you should not focus all your at-tention on running out, or on sinkingthe 9-ball on a combination or kissshot. You should also consider forc-ing a scratch-out, particularly if youropponent has just scratched. Youshould realistically assess the chancesof sinking the 9-ball versus the chancesof scratching-out your opponent.Your decision rests upon how the ballslie upon the table.

Your best chance to play for a run-out usually comes up when the ballsare spread apart. But when balls areclustered, as they often tend to be nearthe beginning of a game, the run-outmay be more difficult, and safety playconsiderably easier. A cluster can beused to your great advantage as a bar-rier, when you shoot the cue ball toone side and the object ball to theother, leaving no clear shot.

When using a cluster as a barrier inthis way, you want to make sure thatit is difficult for your opponent to kickat the object ball. If he can get to the

object ball rather easily by banking thecue ball, you've gained very little ex-cept to keep him from pocketing, andhe might get lucky and leave you safe.In planning your safety shot, imaginewhere you intend to place the cue balland object ball with the shot, and thenput yourself in your opponent's place.If you've left him where it's impossi-ble, or at least very difficult, for himto see a way to hit the object ball,you've made a good choice.

Now for the most important aspectof winning safety play. Remember,you need to force two more scratchesin a row if your opponent has just

scratched, three if he hasn't. So whenyou are planning your first safety shot,you must figure where the object ballwill end up after the shot; it is notenough to merely hide it. The objectball must lie where it will be easy foryou to execute the second safety shot.If you don't have that kind of advan-tageous lie, rethink your first safetyshot until you find a strategy that givesyou a better chance of forcing twoconsecutive scratches.

Suppose the balls lie as in Diagram1. Your opponent has scratched, andyou have ball in hand. Your chancesof running out are not very good —

30 Billiards Digest • June 1990

but notice how the 3-, 4- and 5-ballsare clustered. An effective strategyhere would be to place the cue ball asshown, and shoot the 2-ball to A, leav-ing the cue ball at B. This leaves a dif-ficult kick shot for your opponent. Atthe same time, the 2-ball has beenmoved near another cluster, where youcan execute a second effective safetyby shooting the 2-ball to E and leav-ing the cue ball at D (as in Diagram 2).

You could also use the same clusterfor both safeties by shooting the 2-ballto C, while putting the cue ball at B(dotted lines in each diagram). Afterthe scratch, the 2-ball is shot to F,leaving the cue ball at G (Diagram 2).

If the 2-ball is not near a cluster tobegin with, any nearby ball may serveadequately as the first barrier. For ex-ample, if the 3- and 5-balls wereremoved from Diagram 1, you couldstill hide the cue ball behind the 4-ball,near point B. While it's more difficultto put the cue ball behind one objectball instead of a three-ball cluster, thisshot can be mastered with practice.

Finally, I realize that some readerswill not like the strategy of winningwith safeties. Obviously it's more spec-tacular and satisfying to execute atough run-out. And if you know youcan out shoot your opponent, whybother with a scratch-out strategy atall? For one simple reason: there isconsiderable luck in 9-ball. If you failto run the table, your opponent mightslop in the 9-ball on his next shot. Soyou'd be wiser to go for the win im-mediately by whatever means possible.A game you win through defensive,play counts for just as much as oneyou win by running all the balls.

Some opponents might consider itpoor sportsmanship, even less "man-ly", to win a game by forcing ascratch-out. But remember, you aren'tbreaking any rules; what you're doing,in fact, is taking advantage of one.You're playing smart. You're playingfor "dough," and not for "show."Dr. George Onoda is a research scientist atIBM, and an avid pool and billiard player.

by Dr. George Onoda

Avoiding Miscues On Draw ShotsBelieve it or not, most miscues on draw shotsare caused by hitting the cloth before the ball.

DRAW SHOTS ARE often difficultfor novices and some intermediateplayers. A mediocre draw shot resultsfrom not striking the ball low enough.But when a lower strike is attempted,the dreaded miscue frequently occurs,especially on more forceful strokes.Such miscues are sometimes attributedto a poorly chalked cue, or by theball's being struck too low. But mystudies reveal that the prevalent causefor a miscue is that the cue tip strikesthe table surface before striking theball. I'll show you how I came to thisconclusion, using simple experimentsthat you can try yourself to .confirmmy results.

The first experiment determineshow far off-center a ball can be struckwithout a miscue, using a well-chalkedcue. Although we are primarily in-terested in how low a ball can bestruck below center, it is easier todetermine the maximum off-centerstrike with side English instead, whichshould be the same as for the max-imum distance below center. A usefulfact is that the width of the stripe ona striped pool ball is exactly half thediameter of the ball (Fig. 1). When astriped ball is used as the cue ball, theedge of the stripe itself serves as auseful reference point. With the stripefacing you (Fig. 1), a strike within thestriped area is off-center by less thanhalf the radius of the ball. A strike inthe white area is more than halfwayfrom the center to the edge of the ball.This experiment involves the follow-ing steps: Remove all traces of chalkmarks from the ball. Align the ball sothat it appears as in Fig. 1. Apply aliberal amount of chalk to your tip.Strike the ball to the left of center.Note where the new chalk mark ap-pears on the ball. Repeat these stepsmany times, while varying the strikedistance from the center. Keep trackof where the chalk mark appears aftera successful draw shot and after amiscue.

I have found that the maximum off-center distance that a ball can bereliably struck without a miscue is half

the distance from the center to theedge of the ball (point A in Fig. 1).Miscues do not occur if the ball isstruck within the striped zone; but theyalmost always occur when the ball isstruck in the white area. Try this ex-periment to see if your results are thesame as mine.

Some players believe that the cueball can be struck beyond this halfwaypoint. When the true contact point liesat the edge of the stripe, the center lineof the cue points to the white area.

The edge of the stripe on a ball is half-way between the center and edge of theball. Point A is the maximum off-centerstrike point without incurring a miscue.

This gives the mistaken impressionthat the ball is being struck outside thestriped area.

The next experiment determineswhy miscues occur in draw shots,assuming the tip is well-chalked. Cleana striped ball, removing all chalkmarks. Align the stripe horizontally,as in Fig. 2. Slide the edge of a whitesheet of paper against the bottom ofthe ball, with the paper nearest you.If your tip contacts the ball before hit-ting the ball, no chalk mark will ap-pear on the paper. If the table is struckbefore the ball, a telltale mark will ap-pear on the paper. Try to strike theball at the lower edge of the stripe,with a level cue. When a miscue oc-curs, check the paper and the ball forchalk marks. If the miscue occurredwithout a mark on the paper, see if thechalk mark on the ball is below thestriped zone, which would indicatethat the miscue occurred because theball was struck too low. If there is a

mark on the paper, assuming your cuewas level or close to it, then the miscueoccurred because the table surface wasstruck first.

When I carried out this experimentwith players who miscued frequentlyon draw shots, I found that in mostcases, their miscues were caused by thetips' striking the table first. Only in afew instances could the miscues be at-tributed to hitting the ball too low, in-stead of hitting the table first.

These results led me to understandwhy it is so difficult to strike thehalfway point between the center andbottom edge of the ball. The clearancebetween the bottom edge of the cueand the table surface is quite small.With a cue tip of 13 mm diameter, anda curvature of a nickel, a strike at thebottom edge of the stripe has aclearance of only a few millimeters(1/16 to 1/8 inch) between the bottomof the tip and the table surface (Fig.2). A cue tip that strays downward bya few millimeters during the strokewould strike the table first. The tipwould rebound from the table andstrike the ball with an upward forcecomponent, causing the ball to jumperratically. This effect can be seen byintentionally striking the table justbefore the ball.

From what we've just learned, threemethods for preventing miscues ondraw shots become evident: 1) Use aslightly inclined cue; 2) Use a cue witha smaller tip; or 3) Develop a more ac-curate stroke. For most cases, thethird is the best solution, since the firsttwo have other drawbacks on manyshots; let's examine those drawbacksnext.

An inclined cue increases thedistance between the bottom of the cuetip and the surface of the table, as seenin Fig. 3. This reduces the chance ofstriking the table first. Drawing withan inclined cue is sometimes necessary,as when another ball must be shotover, or when the cue ball is near arail. For the majority of shots, though,a more level cue has advantages overan inclined cue. With an inclined cue,

Billiards Digest • August 1991 27

it is more difficult to aim; it is likelythat the ball will jump (as in a jumpshot); and it is inevitable that the ballwill curve if struck to the left or rightof center.

The size of the cue tip is a factor inthe difficulty of striking the halfwaypoint on the ball. A smaller tip in-creases the distance between the bot-tom of the tip and the table surfacewhen using draw. The clearance is in-creased by about 50 percent when swit-ching from a 13 mm cue tip to a 12mm. But too small a tip wears outmore readily. Also, a thin shaft isbelieved to cause greater squirt (deflec-tion), where a ball struck with left orright English deviates somewhat fromthe aimed direction. We can nowunderstand why tips are not largerthan about 14 mm. If they were, itwould be impossible to strike thehalfway point below the center of theball because there would be noclearance for the tip.

In most cases, the best solutions foravoiding miscues are to use as level acue as possible, and improve the ac-curacy of your stroke. For optimumdraw, the ball should be struck at thehalfway point between the center andedge of the ball without striking thetable first. This requires an accuracyof less than a few millimeters in theplacement of the cue tip on the targetspot on the ball. That accuracy is notso easy to accomplish. For example,if you have trouble banking a ball inthe long direction of the table so thatit comes back to hit your cue tip, itmeans that you are accidentally strik-

A ball struck halfway between the centerand bottom of the bail provides only afew millimeters of clearance betweenthe cue and the table. A miscue occursif the table is struck first accidentally.

The clearance between the cue tip and tableis now much greater. The likelihood ofstriking the table first is greatly reduced.

ing the ball somewhat to the left orright of center by at least a fewmillimeters. The same inaccuracy inthe downward direction would exceedthe clearance dimension of an op-timum draw shot, causing the table tobe struck before the ball.

To improve the accuracy of your

stroke, seek help from a house pro ifpossible, explaining what you want toaccomplish. Or study your stroke witha video system or a mirror. See if youlower the tip as it approaches the ball.Practice developing a smooth,straight, level stroke with a straightfollow-through. Use a short bridge-to-ball distance to minimize the effects ofan upward motion of the cue butt dur-ing the stroke. Make a snug, closedbridge to help keep the cue in line. Atfirst, concentrate on striking the ballsoftly at the halfway point between itscenter and bottom edge, with the cueball around one foot from the objectball. You might be surprised howmuch draw you can get even witha soft stroke, if you successfullystrike the ball at the halfway point.As you become more proficient withthe soft stroke, gradually increase thespeed of the stroke until you can ex-ecute a strong draw shot without mis-cuing.

How well you can draw a ball withconsiderable force is in fact a goodmeasure of the quality of your strokein general. All the knowledge of howballs react on various shots is oflimited value until a good stroke isdeveloped. More attention should begiven to improving your stroke beforeworrying too much about learning theother skills of the game. Emphasizepractice exercises; remember that con-cert pianists began by laboriouslypracticing fingering drills for manyyears.Dr. George Onoda is a research scientist atIBM and an avid pool and billiard player.

28 Billiards Digest • August 1991

by Dr. George Onoda

Coping With SkidSometimes the cue ball grabs the object ball and throws it off line.

Here's why it happens and what you can do about it.

IT DOESN'T MATTER if you shootlike me or you shoot like EarlStrickland, Nick Varner, Mike Sigel orAllen Hopkins; we are all cursed oc-casionally by the dreaded "skid" shot.In a skid shot, an object ball that iscut toward a pocket travels in a direc-tion as if it were cut less thin. At thesame time, the object ball seems toskid (slide) more than normal. Judg-ing from color commentaries invarious taped matches, many top pro-fessionals find skid a mystery. Skidonly happens occasionally, perhapsonce during a long 9-ball set or dur-ing a straight pool match. This wouldamount to around one in 100 to 200shots (as a rough estimate).

I believe that the skid phenomenonhas already been explained correctlyby Robert Byrne (Byrne's Standard

Book of Pool and Billiards, 1987 Ed.,page 129). He calls this phenomenon"cling," but I will stay with "skid"since players seem to use this term fre-quently. Byrne's theory is that skid oc-curs when the normal friction betweenthe ball surfaces is increased. The in-creased friction causes the ball to bethrown in a more forwardly directionthan normal in most types of shots. InByrne's Treasury of Trick Shots inPool and Billiards, pages 152 and 153,he describes several insightful ex-periments by Robert Jewett and BillMarshall. When chalk marks wereplaced at the contact points betweentwo colliding or frozen balls, throw ofcut shots was greatly increased. Sincechalk increases the friction betweentwo balls, the experiments supportByrne's theory.

Figure 2: When the cue ball is struck with high-inside or low-outside, English, there isa chance for a fresh chalk mark to be at the contact point and causing undesired skid.

One or more chalk marks always ex-ist on a cue ball. The mere act ofstriking the ball with a chalked cueproduces a fresh chalk mark on theball. Typically, a fresh chalk mark isaround a quarter of an inch in width.As the ball moves along the table, achalk mark becomes duller andsmaller with time. But it may take ahalf dozen shots before a particularchalk mark completely disappears.Therefore, beside the fresh chalk markproduced by a shot, several older andweaker marks are usually present onthe surface of the cue ball.

If the contact point of the cue ballwith the object ball just happens tocoincide with a chalk mark on the cueball, then enhanced throw can be ex-pected. And if the cue ball has forwardroll, a pronounced transfer of Englishoccurs, giving the object ball some in-itial back spin. The transition fromback spin to zero spin and forwardspin of the object ball is lengthened,accounting for the appearance ofskidding.

Byrne (in his first cited book)suggests that sweat, hair oil and greasefrom french fries could also increasethe friction between balls. But mostevidence indicates that thesesubstances by themselves shouldreduce friction. It seems more likelythat these foreign substances causechalk to cling more strongly to a cueball and last longer as the ball movesaround the table.

Just what are the chances that achalk mark will be present at the pointof contact with an object ball? Aquarter-inch mark covers 0.5 one per-cent of the surface of the ball. If thelocation of one chalk mark is randomwhen two balls collide, the chance thatthe chalk mark is at the contact pointis around 1 in 200 . If other chalkmarks remain from prior shots, moreof the ball's surface is covered withchalk, and the odds for chalk at thecontact point increase, perhaps to bet-ter than 1 in 100.

The greatest likelihood of skid oc-curring is expected when a succession

Billiards Digest • December 1991 25

Figure 1: When the cue ball is struck with high-outside or low-inside English (relative tothe object ball) there is no chance of the fresh chalk mark being at the contact point.

of shots are taken where the cue ballhas traveled only short distances. Thisis because less chalk is rubbed off fromprevious shots, because the totaldistance traveled by the cue ball issmaller. This expectation seems to beconfirmed by watching tournamenttapes.

In shorter shots, the odds that afresh chalk mark will be at the contactpoint with an object ball are much lessrandom. The odds depend on how thecue ball spins as it moves toward theobject ball. When the cue ball is shotwithout side spin, the fresh chalk markcannot come into contact with the ob-ject ball for a cut angle greater thana few degrees. The forward orbackward spin causes the chalk markto remain located somewhere along avertical circle around the center,oriented in the forward direction. Thecontact point lies to either side of thispath and therefore cannot coincidewith the location of the fresh chalkmark. Of course, an older mark couldstill cause a problem.

When a cue ball is sliding with sidespin over a short ball-to-ball distance,the path of the chalk mark is ahorizontal circle about the center. Thisgives it a chance to be at the contactpoint and cause skidding. It can beapproximated (using advanced theory)that a cue ball with maximum side spinwill complete half a revolutionsideways over a distance of around 4inches and one and a half revolutionsover 12 inches. With less side spin, thedistance for these particular revolu-tions become greater.

When a cue ball is struck high andoutside (relative to the cut), it spinsabout an inclined axis, leaning toward

the side of the object ball (Fig. 1). Thesame axis inclination occurs when aball is struck low and inside. A freshchalk mark travels along a small cir-cle, as illustrated in Fig. 1. We see thatthe fresh chalk mark cannot be pre-sent at the contact point. Thus, onlyan older chalk mark could possiblycause skidding for this type of shot.

When a cue ball is struck high andinside, or low and outside, the spinaxis is inclined away from the side ofthe object ball (Fig. 2). The path of afresh chalk mark is illustrated in Fig.2. Depending on the angle of the cutand the amount of English applied,there is a chance that the fresh chalkmark would be located at the contactpoint between the balls.

While we have no control overwhere an older chalk mark will lie ona cue ball, we do have some controlfor fresh chalk marks, which are themost damaging. Summarizing theeffects of spin, a fresh chalk mark willnot contribute to skid when an objectball is cut if the cue ball has no sidespin, if it is struck high and outside orif it is struck low and inside. Bewareof shots where the ball is struck highand inside, low and outside, or whenthe cue ball slides with side English,since the possibility exists that thefresh chalk will cause skid.

In theory, the best preventive for theharm done by an older chalk mark isto clean the ball before each shot. Butin a game, this would hardly be prac-tical. It would slow the game down toomuch and would cause arguments overthe ball placement in cases where avery small error in replacement couldcritically affect making a shot.

A number of practical precautions

could be instigated, however. Wipingthe cue ball with a clean cloth at everypractical opportunity would makesense. Possible opportunities includebefore each game and after eachscratch when the ball is in hand. Also,perhaps in a refereed match, a playercan request the cleaning of the cue ballwhere it lies. The referee would allowthe request if he judges that a slighterror in replacement of the ball is notcritical to the next shot. Anotherimportant precaution, already carriedout in some tournaments, is that ballsshould always be racked by refereeswearing clean, white gloves. Thiswould reduce placing moisture, oil orgrease on the balls.

Two final precautions could also betaken. One is to wash all ballsthoroughly before each day. The otheris to use balls that are relatively new.This reduces the chances of a ballhaving a local rough spot, whichwould also cause increased frictionbetween balls because of its roughnessor because it readily collects chalk par-ticles. If a ball with a damaged roughspot is found, it should be replaced.

The phenomenon of skid is perhapsthe most unfortunate accident in pool.It occurs irrespective of the ability theplayer. It could lose a match for aplayer through no fault of his own.Currently, we just accept it as anunlucky incident that is part of thegame. However, it still seems unfairif it is not the fault of the player. Ibelieve that the precautions that I havementioned should be given seriousconsideration for tournament play.

Dr. George Onoda is a research scientist atIBM and an avid pool and billiard player.

Billiards Digest • December 1991 27

by Dr. George Onoda

Practicing Stun ShotsSimple geometry, using a semicircle, can help

determine if you are executing stun shots properly.

A STOP SHOT occurs when the cueball collides with an object ball head-on and has neither forward spin norbackward spin at the moment of con-tact. A stun shot is a cut shot havingthe identical cue-ball action as the stopshot. Because the cue ball in a stunshot has no forward or backward spin,its path after the carom is a straightline that is perpendicular to the objectball's path. I will refer to this cue ballpath as the 90-degree path. As discuss-ed often in previous articles in thismagazine, the 90-degree path resultingfrom a stun shot serves as very usefulreference for position play.

A stun shot is not easy to executeaccurately for several reasons. Itdepends on three variables; thedistance between the balls, the speedof the shot and the amount of initialback spin applied. It also requires ade-quate visualization of the 90-degreepath.

Proper execution of stun shots re-quires practice. The stop shot shouldbe mastered first, as it will provide asense of the proper action needed fora stun shot. Then practice variousdrills with cut shots, such as thosedescribed in this article.

Because the stun shot involves a90-degree angle, simple geometry canbe used as guide for setting up shotsto practice. A head or foot spot isequidistant to four pockets, two cor-ner and two side pockets. Lines drawnfrom these pockets to the spot form90-degree angles. So we can place theobject ball on a spot and sink it intoone of these pockets. If the cue ballfalls in an adjacent pocket or comesclose, we are applying stun actionabout as accurately as we can expect.For example, as illustrated in Diagramla, shooting the object ball into a cor-ner pocket while sinking the cue ballin the other pocket indicates that thestun action is correct. The same canbe said when the object ball is cut in-to the corner pocket as illustrated inDiagram lb, and the cue ball falls inthe side pocket.

We can extend this idea, throughfurther use of geometry, to find otherpositions on a table that form90-degree angles to two pockets. Weuse the properties of a semicircle,which is a half circle. From any pointalong the semicircle, two lines drawnto the end points of the semicircle arealways at an angle of 90 degrees toeach other. We can now imagine asemicircle extending from one pocketto another on a table and we place the

object ball on this semicircle. Then thelines from the ball to the two pocketsare at 90 degrees to each other. Theposition of the cue ball when it strikesthe object ball is slightly outside of thesemicircle, but to a first approxima-tion, we can suppose that it is essen-tially on the semicircle. Then when theobject ball is sunk in one pocket andthe cue ball falls in the other pocket,you are applying proper stun action.This method can be applied to two

20 Billiards Digest • February 1992

corner pockets (Diagram 2a), a cornerpocket and an adjacent side pocket(Diagram 2b), two corner pocketsalong a long rail (Diagram 3), and twoside pockets (Diagram 4). Wheneverthe object lies along one of thesesemicircles, a cue ball with stun actiontravels toward one pocket when theobject ball is cut into the other pocket.

To use this method for practicingstun shot, place an object ball at a spotor along one of the semicircles. When

the object ball is pocketed, and thecue ball falls in the other pocket orcomes close to it, you are properly ex-ecuting stun shot. Practice with thedifferent semicircles and with varyingangles and distances between the cueball and the object ball.

In the next article, it will be shownhow the semicircle method can be ap-plied in game situations.Dr. George Onoda is a research scientist atIBM and an avid pool and billiard player.

Billiards Digest • February 1992 21