Mark Cramer
C & X Report

You'll find C&X #18 below. Again, some of the text translated kind of goofily -- Mark's word-processing program seems to put a little 'french twist' on some of the text which we are unable to straighten out on this end. The hard copy should be in the mail by the 8th. Sorry about the delay on #17 -- the printer lost the initial mailing list we sent him and didn't let us know for almost a week. We'll try to coordinate things properly with him this time. thanks, Dave



C&X 18

CONTENTS
Editorial: milkshakes
Personal Odds Line Workshop
An Aussie view of Dick Mitchell
Big News about a Little Bet
The End of Science? An Essay
More on the performance box
C&X Caf�

EDITORIAL:
Milkshakes and steroids: dumb and smart corruption
There is legal corruption and illegal corruption. Milkshakes are illegal, and horses testing positive in SoCal have resulted in the quarantine of entire stables. The ancient system of peer pressure has been reversed. Yesterday, trainers felt pressure to use milkshakes. Today, thanks to much embarrassment in DRF headline stories, trainers may feel pressured to not serve up milkshakes to their horses. The penalties have been light, but a precedent may have begun to kick in.
Prior to the recent scandals, a �critical mass� of trainers was probably engaging in dubious race-day procedures. It�s analagous to 80 percent of drivers driving above the speed limit. Comes a time when the cops have no way to enforce that limit. Today, it looks as if the tide could be reversed, and limits will be enforced. Cheating is so endemic that it is assumed by skeptics that trainers and vets will find newer and creative ways to mask illegal drugs or to create new drugs for which no tests exist.
The skeptics may be wrong this time, that is if the peer pressure system finds a way to ostracize cheating trainers.
On the other hand, there exists a quasi-legal type of corruption in the use of steroids during training. In the long term, this trend could be even more disturbing. There are medical pretexts for the legal use of these steroids, while it is certainly known that such medications have other body-enhancing properties than can improve the performance of a thoroughbred. Today�s professional sports records should be accompanied by an asterisk (*broke record during steroid era).
We players have a contradictory role. Under normal market forces, we would stop consuming this product. The stables would then be obligated to engage in steroid-free reform, in order to keep their business rolling. But normal market forces are not prevailing. We players think we�re so smart that we can incorporate the steroid factor into our handicapping. So we keep betting.
Any type of improved performance from a steroid regime during a horse�s layoff would be identified in subsequent improved performances. The pps would tell us what we need to know. In the case of layoff horses winning first time back, we are armed with statistics on the best first-after-layoff trainers, so who cares if they are using steroids or not?
I care. As a player, I would never know for sure which layoff horse had received steroid therapy during his vacation. The player would need a �leading vet standings� to go along with leading riders and leading trainers. As a human being, I would want to know what longterm effects these steroids are having on those magnificent animals who provide us with the greatest entertainment in the world.



PERSONAL ODDS LINE WORKSHOP
The tribulations of handicapping 3-year-olds
Normally I find it soothing for the ego to do workshops on winning races. But good bets are not always winning bets. There�s a lesson to be learned from the 19Feb Grade III Risen Star Stakes at the Fairgrounds, a mile and a sixteenth.
From the get-go there were some serious handicapping obstacles. First and foremost, it was an event for up-and-coming three-year-olds. In such races, past performance is not always as important as projected improvement, as these horses are at a stage in their lives when they will either level out or show great improvement. If you uncover no identifiable reason to project improvement, it might be best to pass such races.
In this race, there was an undefeated Pletcher horse, #10 Harlington, (2 for 2), that had already won at 1 1/8 at GP. At the time he was a Kentucky Derby nominee. Any projected improver would have to get good enough to outrun this undefeated horse.
The second dilemma was dealing with the relative class rank of different home tracks of the horses in the field. The handicapper had to figure out whether Gulfstream and Santa Anita horses had the usual class edge over the local product. You had a Dallas Stewart horse, #3 Storm Surge, eight races, five wins and a place. It was a beautiful win-type performance box. And the horses was 2-for-2 at the Fairgrounds! But were his Fairgrounds races on the same class tier as the SA and GP races of other entrants?
Those were my two most likely winners. I couldn�t separate them.
As I read more deeply into the pps, little by little redeeming traits emerged for other horses in the field, until I had reached the point where it was impossible to eliminate half the field. This meant that I had too many possible contenders to make the line. There was the 2-horse, Electric Light, with a big win at Tampa Bay by 6 ? in an overnight handicap. (I still respect the Big Win method.) There was the 9-horse, Real Dandy, trained by the hot Asmussen, a win and a place in two races at the Fairgrounds, with two wins and four places in eight races: coming from 50,000 Optional Claimers, but having been successful at today�s distance. Maybe not a winner, but those four places looked impressive.
Then there was a Lukas horse, Silent Bid, only one win in six, but it was at today�s distance at one of the class tracks, Santa Anita.
Asmussen had another horse in the field, the 1-horse Smooth Bid, two wins and three places in six races, and finishing close up to Storm Surge, a horse I liked.
This was an attractive field, indeed, and maybe too enticing to bet on. The Pletcher horse began to look like a bet-against possibility. His two wins were in the slop and today it was a fast track. He was leaving from the 10-hole trap. The public was making him 7/5, overbetting him because of the designer trainer. On the other hand, my expected co-favorite, Storm Surge, was going off at 6-1.
There were two logics here. The first one said that in a co-choice race, you bet one of the two preferred horses: the one with clearly the highest odds. Storm Song qualified.
However, there was a contradictory logic. From my own records of betting lines, in races that were clogged with seductive contenders, more often than not, a horse I was not even considering would win the race.
I�m not the only one to have noticed such a phenomenon. When Ed Bain has a race in which there are multiple qualifiers by his trainer stats, he has learned to pass it, for experience has shown that in such cases, a non-qualifier often wins.
Normally the second rule (�pass the race�) should eclipse the first one, but I sometimes rationalize, trying to straddle the line between the two logics, making a small �action� bet on the perceived overlay. By reducing my bet, I rationalize, I am attentive to the essence of the second logic. Sure sounds like I�m flipflopping on this one. Storm Surge looked like an overlay as my co-top choice at 6-1.
Storm Surge raced forwardly, but in the stretch it became evident that the race would be won from behind. Two horses swept past Storm Surge.
Real Dandy, the Asmussen horse that had a 8 2-4-0 performance box finished second, at 45-1. (Another small piece of a huge puzzle that makes the performance box look like a viable factors.)
The winner was a horse that I had not even considered, Scipion, trained by Biancone, showing only a maiden win at 6F from Saratoga, in four races. With AP Indy as his daddy, no wonder he was able to stretch out. In retrospect, it was not such a great stretch of the imagination, but so many others in the field looked so much better.
I bring up this race because it is the most difficult of horsebetting scenarios: the race with many attractive candidates. It�s easy to pass a lesser-of-evils race, but not so easy when the race looks like the display window of an elegant chocolate store.
Scipion had gone from last to first, guided by Gary Stevens. Can you now see how difficult it is to handicap 3-year-old races full of lightly-raced classy horses, all of which have a right to improve.
In retrospect, Storm Surge was a good bet at 6-1. One bad result cannot negate a very good strategy.

AN AUSSIE VIEW OF DICK MITCHELL
On betting two horses to win, and other notes on money management
[Editor�s note. This article orginally appeared in an Australian horseplayer publication called Practical Punting Monthly. A columnist who calls himself �The Optimist� offers a view from down under on the precepts of Dick Mitchell]
Dick Mitchell is a long-time favorite of mine. Here are three of his most useful ideas.
Betting two horses to win in the same race
A principal idea of Dick Mitchell, one I have embraced wholeheartedly, is that backing a second horse in a race reduces your odds by one point, and not by 50 percent. Amongst the purists, the argument goes that if you outlaid $2 with $1 on the first horse and $1 one the second horse, your dividend must be divided by two in order to calculate your correct odds.
Say one horse you like is 10-1 and the other horse is paying off at 15-1. If the first horse wins you receive $11 for your $2 ($22 for your $4 using the $2 unit bet). Your odds are therefore 9-2 or 4.5 � 1. But in point of fact, by taking the bet on the losing $16 horse, you reduced your profit by one point: from 10 to nine. You did not halve your profits, you simply sacrificed a point for insurance. You won 9 for 2, not 9-2.
If the reverse had occurred and the longer priced horse won, then you�d have received $16 for a profit of $14, meaning that once again you had sacrificed a single point for insurance. You did not sacrifice 50 percent. Mitchell makes much of this and his argument is compelling. Interestingly, one of his arguments in this respect is that supporting a second horse can be a lot more effective than betting your top pick win-place.
Mitchell also believes that the bottom acceptable odds for the punter is 3-1, and that below that, you can at best hope to be trading money.
Taking turns
One of Mitchell�s great theories is that horses take turns. If horse A defeated horse B this week, the likely prospect is that horse B will reverse the situation at their next meeting. It happens, and then again, it doesn�t. It is a very interesting point of view, especially if you have two horses that ran 1-2 lining up again, and one is at even money or less and the other is at remarkable odds. Clearly, the discerning player will go after the latter and possibly save on the former, although that may not be possible if the former�s odds are too cramped.
Exotics
Dick Mitchell also says the the only time you should approach a serial bet is if you would have supported the key horses in each leg as straight bets. In other words, you do not allow the track to dictate to you which races you will take your extra risks on.
Betting psychology
He believes that the great discriminator is patience; lose that and you are gone. All your analysis of past performances and betting decisions must be completely objective and not interfered with by temptations.
It is Mitchell who introduced the concept of �comfort zone�. I love that term. It contains a lot of wisdom. Bet within your comfort zone. Anybody who has ever had anything to do with the racing game knows that this is one of the most commonsense philosophies that has ever been put forward. And yet we have all disobeyed it on occasions. That�s when the pulse rate rises, the perspiration starts, and you stop enjoying the whole thing.

BIG NEWS ABOUT A LITTLE BET:
How to cash in on the 10-cent superfecta
Sam Houston Race Park introduced a new twist on a popular wager, beginning Thursday, February 17, 2005, when they offered a ten-cent superfecta on each race. I love this idea, though I can also envision, late in the card, unwary players asking, �Brother can you spare a dime.�
(Another previous Sam Houston innovation has been the pick 3 wager with a 12% takeout.)
Robert L. Bork, president and general manager of Sam Houston Race Park, made the recommendation for the 10-cent-wager in an effort �to encourage new fans to venture into an exotic wager versus the standard win, place or show bet�, a statement which, translated to real English, means: �to increase the handle by sucking in new action�.
"We are still a relatively new racetrack and feel that by offering this highly affordable wager, we will create excitement and attract some new fans," Bork said. "Playing a four-horse superfecta box will cost just $2.40 and the payouts can be hundreds of dollars. We feel that the ten-cent superfecta will be a great addition to the Sam Houston Race Park wagering menu."
Like most racetrack managers, Bork is probably not a player. That�s why he immediately refers to a 4-horse box, not usually the best proposition, instead of referring to the strategy of keying a live longshot, the best way to make money on this type of bet. (A 4-horse box implies that the player believes that each and every combination is of the same value. A key allows you to lean harder on your best overlays, and even to play some winning combinations more than others.)
Sam Houston Race Park got permission from the Texas Racing Commission to add the new wager to its current format. Patrons can make ten- cent superfecta wagers on all races with a minimum of seven betting interests, beginning on Thursday, February 17.
I like this proposition for several reasons. Thinking back to the just-discussed idea on betting two horses to win in a single race, we can be sure that, as a mathematician, Dick Mitchell is aware that multiple bets on a single race, when they include overlay horses, improve the player�s long-term return on investment. The old clich� that playing more than one combination is adding �guaranteed losers� is missing the main point. If by adding a horse, you increase your probability of hitting the race, then you are adding leverage... provided that the horses you add to your combination are being underbet by the crowd.
For example, in an 11-horse field, each horse�s abstract odds are 10-1. But we know that some horses have a better chance than others. Please follow my reasoning.
(a) Suppose that you like one horse that the crowd makes 5-1 (16% chance) and another at 6-1 (14% chance). Combined, that equals a 30% chance.
(b) Now if, by objective handicapping standards, these horses are worth only 8-1 (11%) and 10-1 (9%), then they are underlays, and combined they have a 20% chance, which means you have a built-in disadvantage of 10% in relation to their track odds. Bad news.
(c) However, if the horse the crowd makes 6-1 is intrinsically worth 3-1 (25%) and the horse you like at 8-1 on the totebroad is really worth 4-1 (20%), than you have a 45 percent chance to win, while the crowd, through the odds they produced, only estimated 30%. If the odds you ascribe are accurate, and the crowd�s odds are not, then you have an advantage of 15 percent over the crowd.
If you decided to play only one of those two horses, your chance of collecting would decrease. In the long run, playing one or two horses makes only a slight difference, so long as both are overlays. But the long run maybe many months in the future, and the improved hit rate when betting two horses to win greatly enhances one�s money management possibilities, and also minimizes the chance of Murphy�s Law kicking in: �...that when I like two horses in a race, and choose one of them, the other one will win�.
In exotics, the same concept of multiple-bets-in-one-race is even more valid, for the leverage in exotics increases; you can buy yourself a greater expectation, provided you are talented at picking out worthy longshots at huge overlays.
Here�s why. In a trifecta which finishes with a favorite on top, a 50-1 in second and the second favorite in show, the 50-1 may knock out so many players that your payoff skyrockets. You get extra muscle for an overlay, even when he doesn�t even win!
But HOLD IT: that 50-1 may be worth 100-1. You have to know some reason why the public is so far of on such a big-odds horse. You need the 50-1 to be worth no more than 8-1 or 9-1, and if you use him in the top slot, then you want his intrinsic worth to be no more than 6-1.
In other words, you needed to have a huge insight into this horse, knowing things that the crowd did not know. Or, that 50-1 may have had a running style conducive to finishing second, with little hope of winning. With the exotics pools so much more complex than the win pools, radical overlay combinations are more easily found. That�s why many (but not all) professional players concentrate on exotics, even though the takeout is higher.
That is also why Susan Sweeney can afford to use the ALL in an exotic, provided she has overlay insight into the other positions or legs of the exotic, and at the same time, has identified a chaos situation where the ALL is likely to be remunerative, such as a false favorite or totally wide-open race in which lessers-of-evils face off.
This is where the 10-cent superfecta comes in. This wager allows the smaller or moderate player to compete with the big boys. The high rollers will lose their leverage advantage. I have witnessed conversations among big Pick 6 investors who decided to stay out of the $1 pick-6 pool at Canterbury Park, �...because with a dollar bet, the competetion can afford to go deeper, and we lose some of our advantage.�
I would suspect that, in this same logic, large players who enjoy a volume advantage would not like the 10-cent super at all.
The 10-cent superfecta would allow you to do all kinds of things that might have been beyond your comfort zone: like using the ALL in the fourth spot of the super in races where any and every pace scenario is possible, effectively converting it into a trifecta. The concept is that the best horses in a race will be fighting it out, and some of those best horses will �die� in battle before the wire and finish out of the top four, leaving third or fourth place for a longshot that can pick up the pieces. Another possible scenario is the longshot that has an early pace advantage and can hold on to third or fourth only because he�s been able to enjoy a slow early pace.
The second reason for appreciating Bork�s innovation is that this type of wager is what �pari-mutuel� is all about. The French invented pari-mutuel as a way to share the action (�pari� is bet and �mutuel� is share) in the tradition of �galit� (equality). In the win hole, everyone gets the exact same odds, whether they bet two bucks or two hundred. Yet, in exotic wagers that require many combinations, players with huge bankrolls have an advantage over others whose comfort level is with a lower investment. A bachelor with no kids to support and a rich lady friend can bet all he wants without reservations. A father of six with a wife who doesn�t work will have to think twice about putting $96 into a single superfecta with a longshot as key. He knows he has a good bet, but he also knows that he could easily lose ten of these consecutively. But if he could do the same bet for $9.60, he�d still be in for a handsome return if his longshot comes in any of the top four spots.
It�s the oft-referred-to idea of Barry Meadow, that it�s better to have a small share of a large pick six ticket with competent but not brilliant handicapping, than have all of a $48 ticket with brilliant handicapping. Until now, a heavy better with reasonable intelligence had a huge advantage over the most clever handicapper with a small bankroll. In other words, everyone was not getting the same odds. With the 10-cent super, everyone essential can play with the same odds.
You may recall a few years ago that I wrote an article about the dilemma of the �if horse�, the type of horse that you cannot eliminate but that becomes too expensive to include in an exotic bet. With the 10-cent superfecta, provided that the handicapper has made some real discovery about the race, he could afford to include the �if� horses. That way, if his 88-1 longshot finishes third, he will not lose the super because of not being able to use several if horses in the fourth slots. (I once got a 95-1 horse home first, had the place horse, and missed the tri because I did not use all my �if� horses in the third slot.)
I have discovered the extraordinary power of the 10-cent superfecta three years before the bet was invented, when they initiated the �Multi� wager in France. In the Multi, you have to pick the top four finishers, in any order, in fields of 14 or more horses. For the price of one four horse wager, you could buy a six or seven-horse combination. Six horses meant 15 combinations. Seven horses meant 35 combinations, for the price of one. I gradually discovered that it was better to do several combinations with five key horses, adding several sixth horses on different tickets. Unless I could comfortably throw out the top two faves, I discarded the 7-horse combo, because it gave the same weight to my seventh choice as my first choice. Getting 15 combinations (six horses) for 3 bucks in France is equivalent to twenty 15-cent wagers.
Most important, I had the leverage to include if horses.
This has been my best wager ever, in long-term r.o.i. as well as in consistency of hits. However, it has been successful for me because I remain selective, only betting when I feel I have some special insight. That�s the problem with the 10-cent super on every race with 7 or more horses. This wager could very well attract lots of dumb money.
We must remember that betting professionally has little to do with how much you bet and much more to do with the way you manage your betting, as Steve Fierro has pointed on in a previous C&X. The 10-cent superfecta is a great bet for a good longshot player who feels he can key a live longshot and maybe demote a favorite at the same time.
We must avoid the nightmare scenario, where the player says, �hell, it�s only 10 cents�, and adds horses for no reason but to have a bigger ticket, doing this race after race. This prototypical player feels he can somehow get away with less thinking, because he�s betting less money. Betting tragedy is inevitable under such circumstances.
It�s the great trap, in an otherwise exciting new bet. If readers are interested in this bet, I would suggest following the horses at Sam Houston on a regular basis, as if it were your home track. There�s no edge for an outsider trying to outsmart the locals. I�d say that it would be smart to get to know the local Sam Houston horses as if they were all in your own personal stable.
A professional treats all bets with the same level of care and thought. The mix of cheap bets and sloppy thinking is a brew for disaster. But 10-cent wagers in the hands of a discerning player who knows the value of leverage can open the gates to a new realm of profits.

THE END OF SCIENCE?
An essay on the Age of Enlightenment in handicapping literature and the validity today of past discoveries

When Dr. Howard Sartin began writing about �fractals� in the 1990s, it was a sign that the literature of horse race handicapping had passed into a different phase. The discoveries that had begun in the late 1970s, based on scientific observation and research, had reached a subatomic or extragalactic stage, where broad concepts of knowledge seemed to have reached the outer edge, and it was increasingly difficult to travel beyond the galaxy of handicapping factors.
In this essay, I would like to trace the development of the major breakthroughs in handicapping, in each case, commenting on their relevance today. I will not be governed by strict chronological order, but rather an intellectual order of things.
For this reason, we begin with Andrew Beyer, from his Picking Winners. We all know about his speed figures, and the fact that they appear in bold print in the DRF does not give us a great information edge. But the concept behind them, track variant, remains vibrant today. Variant, of course, is a figure attached to the relative speed, not of horses but of the track itself. A 1:10 on a lightning fast track is not evaluated the same way as one on a sluggish track.
I have found two uses for track variant today, beyond the Beyer fig. When I discover two horses coming from different races, same distance and same day, I apply what I call �the instant variant�. If those two horses are entered in the same race, it does not help me too much, for every player will see the comparison. However, if they are entered in different races, I may discover that the 1:10.2 from the running line of a horse today entered in an optional claiming race is especially valuable in the context of a 1:10.1 from a different pp, of a horse that is today entered in a Grade 2 sprint.
Happy Hooter 1:10.2 from Race 6, 16JulyMon, entered today in Optional Claimer
Grungy Gump 1:10.1 from Race 8, 16JulyMon, entered today in a Grade 2
In judging the chances of Happy Hooter, the handicapper would wonder whether that 1:10.2 was on a lightning fast track or on a fair racing surface. By noting that the same day, same surface, same distance, the Grade 2 horse Grungy Gump, did a 1:10.1, it makes Happy Hooter�s time look better. If I were to find a third horse�s time from the same day, I could triangulate my view of the final times, relative to the class of race they came from.
In other words, if you see an apparently fast time, you need to scan the pps for the times of other horses that raced the same day, same distance, and then you can view the time in the context of the day it was earned.
Another use for track variant is in foreign tracks around the world, since no one calculates the variants at those tracks. Of course, you have to be there to do your variants, but this is a sign of the universality of Beyer�s great discovery.
During a similar period, we found James Quinn�s Handicapper�s Condition Book on the shelves of studious players. Emerging from this great book was the notion that the conditions for most races favor lightly-raced horses over those with many more races for the period listed in the conditions: 6 months, year, lifetime, etc. This Quinn discover is even more valuable today than it was back then, for with year-round racing, you find more over-raced and stale horses that can be beaten by layoff comebackers or horses with only one or two preps since their time off.
Without necessarily thinking of Quinn, Ed Bain rediscovered the wheel, for his own profit, when he decided to focus on horses that had no more than four races since their layoff. Of course, Bain correlated the lightly-raced factor with the trainer specialty, also a derived idea originating with Steve Davidowitz from his Betting Thoroughbreds, though to Bain�s credit, he believed nothing and started all his research from scratch.
Another discovery of Davidowitz was the key race. It was a laborious process involving keeping of results charts and circling any horse from in chart if it had come back to win its next race. If a particular result chart accumulated circles, you had a �key� or �productive� race. For a time, the DRF began publishing key races, but then they dropped it and instead put the names of horses that came back to win next outing in italics at the end of a running line. Those italics are limited to the top three finishers from a race, but with the original key race procedure, you would know if horses from the back of the field had come back to win in their next race.
Today, the concept certainly retains a certain degree of validity. But I learned that it was even better to be one step ahead and not wait for circles to accumulate in results charts. Therefore, I note down when a particular race is full of horses that seem classier than the official class designation of the race, and as Quirin fig people still do, I note immediately if a race seems to have been run faster than an �instant� par, by comparing its time with other races run the same day.
That transitions us into William Quirin, who along with Beyer, developed the idea of par times. Par times are essential if one is to calculate any type of figures, for we need to know the relative �speed� of different race tracks and of different distances at any one given track. This allows us to compare different tracks or different distances.
You can purchase par times from a number of sources, with Gordon Pine�s pars being among the most readable. However, the handicapper can do �instant pars� just by keeping notes of the times at the various tracks. Now that multi-track simulcasting is engrained in racing culture, it doesn�t take long to discover with Philadelphia or Laurel are faster tracks during a particular season.
For me the greatest Quirin discovery was the notion of attaching a return on investment to particular factors, and especially appealing were the list of Quirin turf sires with a neat r.o.i. next to each sire. The problem was that the shorter the sample of a sire�s progeny, the easier it was to show a profitable r.o.i. and you needed to attach an asterisk to newer sire records.
Quirin�s turf sire records eventually led to Tomlinson numbers, though without the r.o.i., which for me were a regular source of signers in the 1990s, before they were published in the DRF.
There were periods of time when my wagering was dedicated exclusively to turf maiden races and allowance events with horses that had never been on the grass. The return on investment was so much greater than any other factor that I stopped bothering with other races.
I later incorporated the Mike Helm sire ratings in various categories, including turf, debut sires, stamina index and class. Helm�s stuff is still as good as it was in the beginning, and has the advantage that it�s not published in the DRF, while Tomlinson is.
As I�ve written before in C&X that it�s tougher to exact a profit today from turf sire ratings for two reasons: first, that they are published for everyone to see; and second, that trainers today are much more aware of pedigree and less likely to enter horses on the grass when these horses project to hate the grass. That means that it�s tougher to find a maiden turf field in which only one or two horses qualify on pedigree.
(As you already know, in France I use the reverse-Tomlinson method, betting American bred-for-dirt horses switching from turf to dirt.) Back in the USA, the Northern Dancer line continues to be the most productive on grass, and this info is especially useful with young sires that can be traced to that line.
Back to Quirin, the idea of attaching a return-on-investment figure to a handicapping factor has been enormously useful for research, and will continue to be so. Less useful was the great discover by Frederick Davis called impact value, later incorporated by Quirin. I.V., as it is known, simply tells you whether a certain factor wins more or less than its fair share of races.
If grey horses account for 1 percent of all entrants but win 2 percent of all races, then they would have a 2.0 I.V. But if greys accounted for 4 percent of all races and only won 2 percent of the time, they would have a negative 0.5 percent I.V.
I felt that I.V. had become a distortion in Quirin�s literature because you could have a factor with a positive I.V. that shows a flat-bet loss (because it�s overbet) and one with a negative I.V. that shows a flat-bet profit (because it�s underbet).
For me, the r.o.i. concept of Quirin is everlasting. One only needs to understand that the shorter the sample, the less likely that a positive r.o.i. is sustainable over the long run. On the other hand, a short-term positive r.o.i. could mean that a particular factor is hot now. The DRF, of course, now uses r.o.i. with its trainer stats.
Another realm of scientific discovery in handicapping concerned Ragozin�s �The Sheets� and a couple of offshoots including Thorograph. The Sheets did two new things at the time. They graph performance cycle, something I had been doing long ago, but which the Sheets do more effectively, and they have a new system for creating speed figs. The fact that the lower the number the faster the fig in the sheets is just a gimmick, but the clarity of using simpler figures is strikingly useful.
In fact, the great thing about the Sheets is the visual format. It has been known for quite some time that human intelligence can be divided into more than a few categories (I�ve seen the number 8 somewhere, but that will certainly increase as brain research becomes more sophisticated). One of those categories is �visual intelligence�. People who are visually oriented can truly appreciate the format of The Sheets.
It was Ragozin who first came up with the notion of �the bounce�, and to their credit, most Sheets people do not apply this concept in simplistic ways, since not all animals have the same physical dynamics.
But there is a universal validity in the idea that an athlete who overperforms after a vacation can come back a little sore and may need time to work through the soreness. In France it�s known among trainers as �the famous second race back�. However, as with the Tominson numbers, American trainers are increasingly aware of the bounce possibility (many use The Sheets) and take measures to avoid the bounce, by not running a horse right back in its second race after a layoff, or by racing the horse lightly in its first race back.
In any case, with the Sheets, the player can see a pattern of performance and visualize which way a horse is going. This requires a certain degree of artistic interpretation. Two Sheets guys could have two different interpretations, and I believe this is a strength and not a weakness of The Sheets.
Where I do find a flaw is within the methodology of carving out Sheets figures, in which an adjustment for the time figure is made by incorporating the extra distance a horse runs when forced to race wide. In principle, the concept is beautiful, but applied mechanically, it fails to incorporate the fact that it is often better to be in the middle of the track than on the inside, depending on the bias, but also depending on where the least traffic is. So a particular Sheets fig may demote one horse�s number because he had the rail all the way and inflate another�s because he was wide, even though on that particular day or in those particular races, the rail was slower and the horse had raced in tight quarters along the rail.
Beyer has asked this question to Sheets people and I�m not sure he�s ever gotten a straight angle. Within the sheet community, there are intellectual differences between Sheets and Thoroughgraph about such things as wind velocity, which for me are wasteful arguments, since so many imperfections already exist in a final figure that trying to make them �more� perfect by incorporating a questionable factor does not seem like the most advanced science to me.
That is not to say that wind does not play a role. I know this well, since I am a bicycle commuter. Even a moderate wind effects my �final arrival time�. But wind is not usually steady enough to calculate. It can whip up during one part of a race and calm down during another, and only the guy riding the horse is in a position to tell you.
You must have discerned that I agree with Beyer in his question about the �wide� measure in Sheets figs. Beyer has been acutely aware of track bias, and a reading of his My $50,000 Year at the Races shows the reader that the discovery of a real bias, especially at Pimlico, was in great part responsible for Beyer�s profits that particular year.
Bias comes and goes and is hard to get a handle on. If you suspect the existence of a bias, you can cash in that very day, or you can wait until the horse that raced against the bias comes back on a fair track.
However, on a given day, if front runners win all 10 races, and the average mutuel is only $4.40, then you do not have a bias; it is simply coincidental that the best horse in each race happened to be a front runner. On the other hand, if front runners are winning races with an average mutuel of $33.00, then something is happening on the track or within the mass psychology of the jockey community. Of course, the converse bias would be true if closers are winning most races with a high average mutuel. Ditto for rail horses as compared to wide horses. I�ve seen days when one longshot after another won races by going wide. A good jockey sometimes tips us on bias, by staying away from the rail on purpose when he doesn�t have to.
There�s much more to say about bias, but the mention here is simply to relate to yet another Beyer discovery.
The issue of track bias was taken to its most exquisite level by Dr. Howard Sartin when he taught his followers to develop track profiles. In essence, certain tracks at certain distances favor certain types of runners. The Sartin methodology was based on pace, and derived from Huey Mahl�s The Race is Pace. Using the concept of energy spent, Sartin and his colleagues, including Tom Brohamer, developed subtle pace programs based on feet per second.
With a typical Sartin program, you got to see, visually, how certain horses expend comparatively more energy in the early going while others are more �sustained� and keep more energy for later in the race. In fact, all American dirt Tbreds run faster earlier fractions than later fractions. Even a horse that is coming from behind in the stretch is actually slowing down; but he�s slowing down less than the other horses.
Brohamer gave us a simpler way of seeing this, by categorizing horses as E (early), P (presser) or L (late). Essentially, all this stuff was basic pace handicapping becoming more sophisticated. Sartin followers were left naked when trying to handicap a grass race and for them, even routes were less dependable than sprints, but I found the Sartin method intellectually stimulating.
In any case, with Sartin and Brohamer, we learned that it�s not so simple as playing the closer when the field is full of early speed, or playing the obvious lone front runner. What if you have five need-to-lead horses and one of them projects to get the lead. That means that the other four suddenly become follows, right out of the gate, and the speed of the speed wins.
To this very day, you the handicapper can benefit by reading past performance lines on the basis of energy spent, and Quinn and others have noted that in turf racing, it�s the sustained (read �late�) energy that often wins.
Getting back to the notion of different intelligences, one type of intelligence is pattern recognition. But forget about so-called I.Q., for pattern recognition can be learned and is not innate. I�ve done a lot of work in this area, and I would venture to say that this is one of the most dynamic realms of horse race handicapping today. If you read the good work of many of today�s fine turf analysts (I mention Dave Litfin and Brad Free, but I could mention 15 or 20 other names as well), you will see that they are recognizing patterns in a way that combines science and art.
In the end, the scientific discoveries since the 1970s must be interpreted artistically. We are not at the �end of science� but for the time being, much of the newer stuff is derived from the above luminaries and sometimes carried to a higher level.
The quarks of horse race handicapping may be yet undiscovered. But we are at a point where it is apparent that the science of the past is merging with the artistic interpretations of the present. I hope I�m not forcing the jazz metaphor, but I do believe it applies. The above discoveries and principles are like a great classic songs. But to win, you need to be a John Coltrane, extracting the hidden juices from My Favorite Things. Most folks can learn to sing or play the basic melody of the past performances. But rich harmonizing and synchronicity are what lead to the profitable extra we need, since the betting public is partly in tune to the science of the past three decades, in part thanks to the innovations of the Daily Racing Form.

MORE ON THE PERFORMANCE BOX
Letter from a reader
Mark
when using the performance box to find a horse with 25% winners and more wins than combined places and shows, what is the minimum number of races required for evaluation?
Bill

Hi Bill,
Good question, and I regret that I have no mechanical answer. Certain extremes are easy. Ed Bain would never play a horse that�s 1-for-19. But in most cases, as you can discern from the previous article, it�s a question of artistic interpretation. A horse that�s 2-for-19 lifetime, but 2-for-3 at 7 furlongs might be a clever bet at 7f.
Considering that lightly-raced horses have an advantage in today�s races, a horse that is 3-for-5, or even 2-for-4 could be a good bet, even though it�s a short sample. You�d have to look at the pps to determine how these victories were earned, just to confirm, visually, that they are indeed �win types�.
I think that the overall performance box is a question of pattern recognition but in combination with the dynamics of today�s race. A horse with a 23/1-9-6 sure looks like a partial backwheel candidate, but what if he�s a deep closer, and today, the track is so much in favor of early speed that he has no chance to make progress in the lane?
On the other hand, a horse that�s 7/1-4-0, a smaller sample, might be a better play as a backwheel, depending on the dynamics of the race, and the reasons in his past races for earning such a record. If he has been racing above his proper class level in most of those seven races, then he may not be a hanger at all. The telling sign for using a short sample for a backwheel is when the horse has earned the high proportion of place finishes in both higher and lower class levels, thus nullifying the class variable.
Sorry for the cop-out, but the performance box, even when used as a primary factor, is subject to adjustments like any other handicapping factor.
That said, I love this factor, and it�s especially telling when the you isolate the performance (1) at the distance; (2) at the track; or (3) on the today�s surface.
Let me give you at least one example. Not a text book example but one that gives an idea of how to interpret information related to the performance box.
The San Carlos, Grade II, SA, 7f, 26Feb05
In this 8-horse field, with little early speed, there were three easy eliminations, two of them that appeared to be win types. The 8, a Brazilian horse named Oceanus, was a deep closer that was outclassed, coming from a best level of 62,000 claiming. The 3-horse, Choctaw Nation, was 3-for-3 for Jeff Mullins, but all in route races. (Since Mullins horses were placed under quarantine, I suspected there would be a decline in his high win percentage.) Choctaw Nation had five wins in six starts, but four of those wins were in routes. His only sprint race was at 6 ? furlongs as a first-time starter, and came as a deep closer. The 7-horse, Del Mar Show, from the Lukas barn, was another win type: 35/11-5-3, but had never won a sprint and was trying the dirt for the first time. He figured to not have enough speed.
With no quick early speed, a �race shape� had materialized. With a pace advantage, the early horses could not be expected to back up, so the above-three deep closers could be safely eliminated from the win and the exacta.
There were two other win types. The favorite, Mass Media, was 9/5-1-0. The 7-2 Hasty Kris was 14/5-1-1 at today�s 7-furlong distance. These were the only two win-types that could handle the distance in this particular pace scenario.
There was only one place type, but you could not see it clearly in his distance performance box, where he 1/0-0-1: Harvard Avenue. However, HA�s third-place finish came in the Grade I Malibu at 21-1, where he finished in front of Mass Media, and less than a length behind the winner, Rock Hard Ten. Though he was a come-from-behinder, his up-close finish at today�s 7f distance at Grade I, gave him credentials to handle the countercurrent. If you looked at his performance box for Dirt-Fast, he was 10/1-4-3, giving him a distinct cha nce for second in the exacta, but making him an unlikely winner. As a classy closer in a field that did not favor closers, he was the type of horse that might come up a little short.
And that�s the way it was. The exacta with the 7-2 Hasty Kris winning on top of the longshot Harvard Ave paid $33.80 for a buck.
Performance box was the key to deciphering this particular race, but it needed to be interpreted in the context of the pace and class dynamics of the race. Harvard Ave was racing against the pace current but had a class advantage, given his Malibu finish.
One last note. Bill understands, and the rest of you should also understand, that I have my file of losing races in which the performance box did not work out so effectively as a handicapping factor. Space does not allow for critical analysis of a losing race, but believe me, with any factor, we should always do critical analyses of losing races so that we can develop greater precision as to when to use and when not to use the factor. Performance box is not a perfect factor, but in the long run, it functions well, since it is underused by the public.

C&X CAFE
Mark,
I wanted to ask you a question about the elimination rules in the �short form�. I have been seeing a lot of races being won by trainers and jockeys in the 5% range. I wanted to get your thought on this happening over and over again. Please let me know what you think.
Thanks,
Ron

Ron,
You say that you've seen "a lot of" races won by 5 percent trainers or jockeys. What do you mean by "a lot of"?
In a sample of 1,000 races, 50 will be won by 5-percent trainers. That seems like a lot of races. But in the long run it�s not. Of course, if you have a trainer who�s 1 for 20 this year (5 percent) but was a 15 percent trainer last year, then common sense tells us that the 1-for-20 is too small a sample.
In most cases, five-percent trainers will win five percent of their races. Would you like to collect only 5 percent of the time?
Try to think in percentages and not in vague, impressionistic terms like "a lot of". You might see a particular card in which six of the nine races are won by 5-percent trainers. In the realm of statistics, that is entirely possible. But in the long run, a 5-percent trainer wins five percent of the time. If he gets hot, which would only be for a temporary period, his percentage could double, to 10 percent. Would you be happy winning 10 percent of the time?
Cheers,
Mark

From Bob
I had a nifty 8-1 pattern match at Philadelphia friday. �2nd race. �Harry Vega riding for a 0-49
trainer. �0-49 is an automatic elimination. �BUT, what's a top jockey doing on this horse? �I noticed Vega had ridden the horse before under similar circumstances. �Del to Pha---and now Pim to Pha.... It was almost a triple match at 8-1. �The exacta was $77.00

Cramer responds:
Here you have evidence of the power of the pattern match (see final part of THE END OF SCIENCE?) and how there�s no one formula for identifying such matches. This is an example of the art of handicapping, based on science, but with crafty improv.