After showing a flat-bet profit on the totality of his full-card selections over a series of Southern California race meets, why did TimeformUS West Coast Analyst Justin Finch begin writing the longshot-themed “On The Contrary” column? This is a barely edited transcript of a Q & A we (TFUS) conducted with Finch (JF) on the relative merits of handicapping and wagering on longshots. Questions or comments? Please drop us an e-mail at firstname.lastname@example.org, or by all means use the feedback section below the piece.
To search the archives for On The Contrary results, click here.
TFUS: We’ve been surprised at the reaction to On The Contrary. On the one hand, there’s been a big audience and enthusiasm. On the other, we’ve gotten feedback from customers who assume a focus on longshots means that a horseplayer is desperate or reaching. Does playing longshots equal desperation?
JF: The answer to this question is in the betting records of the person doing the playing. Do some people reach for longshots out of desperation and have a terrible ROI with them? Of course, and these people should be very wary when they find themselves liking a longshot. Longshots are not for everybody. So much of the game is psychological, a matter of knowing oneself. For example, most big longshots look awful on paper. Now if you have the type of personality where you sometimes find yourself liking something about a horse who looks awful on paper, and you bet this longshot, and he loses badly, as longshots usually do, and then you start beating yourself up (“God, I KNEW that horse had no chance; he was totally overmatched on paper. Why did I waste my money on him?”), and you get thrown off your game, then you should not bet on longshots. It is as simple as that. The only people who should be making a steady diet out of playing longshots are people who have trained themselves to look past the inevitable strings of losers and focus on the hoped-for profits at the end of the line. And even then, if one’s betting records do not justify playing longshots, one should not play longshots.
TFUS: Do you have anything more to say to the people (and we hear from lots of them) who say playing longshots is the worst thing you can do because it automatically equals desperation?
JF: Yes. To quote Nietzsche, “The so-called ‘worst things,’ perhaps they are merely the best slandered.”
TFUS: OK, but what is “Gambler’s Ruin” and what is its effect on longshot players?
JF: The term has taken on different meanings over the years, but for our purposes, “Gambler’s Ruin” essentially means that you can play this game with a proven mathematical edge on each bet and still have your bankroll disappear because of the mathematically guaranteed losing streaks that will come, and this is going to be particularly true when playing longshots. Gambler’s Ruin is real. It can be proven mathematically or through computer simulations or even by sitting at a table, throwing dice, and recording the results over a huge sample. It is extremely important to guard against it. But you can guard against it by choosing bet sizes that don’t threaten your bankroll, by reducing your bet sizes during periods when your bankroll gets thin, by mixing in bets on horses at lower odds, and by diluting your longshot bets by using your longshots underneath in exotics and putting less money on top. It is very important that people do this. If they don’t, they could end up in the horribly frustrating position of having worked hard enough to acquire an edge in this brutally difficult game while having no money to show for it. It would be as if they invented the wings to fly, only to watch someone take away the sky.
TFUS: Playing longshots seems to have fallen out of favor a bit over the years, hasn’t it?
JF: Yes, and you can see that if you look at the ROI spread in different odds ranges. There used to be a big spread between favorites (which lost maybe half the takeout collectively) and longshots (which lost considerably more than the takeout). The last time I saw stats on this, the spread had withered to essentially nothing, which itself makes longshots more attractive than they once were, or, if you prefer, less unattractive. I wonder if what causes a lot of this anti-longshot talk is chalk players who themselves play longshots only when they are deep in the hole and feeling desperate. Perhaps these chalk players then project their own desperation onto others. Yes, if you’re enthralled by some handicapping book that is full of senseless rules and mindless, dictatorial blather about how you should play only horses who have a checklist of factors in their favor, or only horses who “fit the conditions of the race,” or only horses who are proven at the distance, or only horses trained by good trainers, etc., etc. etc., then every longshot is going to look like a desperation bet because every longshot is going to look hopeless. But one needs to consider the source of this “desperation” talk. There’s a great scene in the Coen Brothers movie “Burn After Reading.” [Disclaimer: It’s a comedy, and it means to give no offense.] At one point, Malkovich gets called into his superior’s office and told that he has a drinking problem. He responds, “My drinking problem? You’re a Mormon. Compared to you, everybody has a drinking problem.” Every day, I see (or at least I think I see) numerous races in which multiple horses are going to post at similarly high odds and I am fairly certain that one of them is prepared to run vastly better than the other. One might be 20-1 and I think he deserves to be 10-1. One might be 20-1 and I think he deserves to be 80-1. I’m sure I’m wrong about these assessments plenty, but I am right often enough to say that there is something interesting going on here–some real inefficiency at the high-odds end, certainly more than I see at the low-odds end.
TFUS: So there’s no demonstrable edge in excluding all horses that go off as longshots?
JF: The overwhelming majority of people lose at this game. The track takeout is onerous beyond belief. The game is brutally hard to beat. The reason people lose over the long haul is that they are putting too much money on underlays and not enough on overlays. To me, the idea of trying to beat a game this tough while taking an entire class of potential overlays (longshots) out of the equation is preposterous.
TFUS: What do you think has caused longshots and favorites to even out over the years?
JF: Market corrections, increased efficiency in the market, whatever you want to call it. The crowd is no longer overrating the chances of longshots as a group by a significant amount. But that really has little to do with my point here. It seems to me that increasingly, people think that one way to sound “sophisticated” is to badmouth longshot players. Yes, there are sound mathematical reasons to take care with longshots. Gambler’s Ruin is one of them. Keynesian “In the long run, we are all dead” logic is relevant as well. These are legitimate concerns. Longshot players need to keep clear of the sort of bankroll recklessness that opens them up to these dangers. But, having accomplished that, there is nothing inherently wrong with playing overlaid longshots and a lot inherently right with it.
TFUS: Is there anything inherently wrong with playing favorites?
JF: No, but any horseplayer who keeps his eyes open will see that horseplayers are everywhere tapping out on low-odds horses, too. These people are playing low-odds horses without an edge. And soon they are not playing the horses at all. And I believe that one of the reasons they are tapping out is that they just plain do not love the work that they are doing. Who the hell wants to wake up early every morning and grind away on [expletive deleted] 9-5 shots who look like overlays but quite possibly aren’t? For these people, making an attempt to build an approach around longshots might well prove to be positively rejuvenating. What is more, the things they uncover might help them with 9-5 shots, too. One thing that for me is too painful to contemplate is the idea of trying to play this game well without being excited about waking up every morning and handicapping horse races.
TFUS: We recently read this piece at FiveThirtyEight.com on the “illusion of causality.” It’s fundamentally easier to convince someone that they’re playing smart when they bet a 9/5 shot whose fair odds were 7/5. But they’re far less comfortable with betting a 30-1 shot whose fair odds are 10-1. Assume that one race has a 9/5 shot as described and a different race has a 30-1 shot as described. Which one is a better bet and why do you think there’s such a strong instinct to settle on lower-priced horses?
JF: Assuming that the “fair odds” are accurate, my feeling is that both of these horses should be bet, but the 9-5 shot should be bet more heavily even though the 30-1 shot is the better bet in a pure, idealized sense. What I mean is, in the long run, the 30-1 shot will produce higher profits, but in the long run, we are all dead. But, again, I would absolutely bet on both horses. But his line of questions triggers several in return. Do you think we will get the 9-5 odds when the race is over, or will he get hammered down to 7/5 as the gate opens or as the horses round the far turn? How about on the longshot? Do you think we will get the 30-1? Which one is more likely to go off higher? Do you think this stuff balances out? I don’t. As for why so many people have a strong inclination to settle on lower-priced horses, we all know the feeling when our bet loses. The heart sinks. For some people, it sinks for only a few seconds. For others, it remains sunk for a long time. Regardless, to me, the real question here isn’t the one you posed above. The real question is why someone would settle for an underlaid 7-5 shot over an overlaid 30-1 shot. To do that is to consign oneself to the poorhouse. It is positively sinister.
TFUS: Any other comments on the illusion of causality?
JF: The subject is highly relevant to betting on racing and so important that it deserves treatment that goes well beyond what can be accomplished in a short article. The article you linked to is from Nate Silver’s website. Silver wrote a book called “The Signal and the Noise.” It is, I think, a terrific book for handicappers to read. So is Daniel Kahneman‘s “Thinking, Fast and Slow.” Kahneman is brilliant and modest and endlessly endearing and he’s got a fine discussion of “base rates,” which handicappers can profit from in a myriad of ways. But I feel that the best books of all, for handicapping purposes, are the books by Nassim Nicholas Taleb. “Fooled by Randomness,” “The Black Swan,” and “Antifragile” are loaded with ideas and observations that I consider indispensable. Suffice it to say that if I had a child who wanted to be a good handicapper, I would make her swim in these books for months before I even taught her how to read the PPs.
TFUS: How are Taleb’s ideas relevant to horseplayers and playing longshots?
JF: Taleb is not a horseplayer. He came to prominence as a trader in the 1987 stock market crash. I take from him not so much WHAT to think but HOW to think. Call it an attitude toward the world. From Taleb I learned to adopt an attitude like this one: We walk around all day thinking we know things when in fact we don’t. We think we see causes and effects when in fact we see randomness and nothingness. We think we have things all figured out when in fact the world is about to crumble beneath our feet.We think highly of ourselves because of our so-called knowledge when in fact we ought to be cursing ourselves because of our ignorance. We see a handicapper who is red-hot and we think he has the game figured out. What we don’t see are all the handicappers losing while employing the exact same methods. We think we are seeing meaningfulness when in fact we are seeing meaninglessness. We see a horse improve with blinkers second time out and we think the blinkers caused the improvement when in fact they probably had nothing to do with it. We see a horse improve the first time he races on grass and we think we are seeing a grass horse when in fact we are seeing a horse who just improved for reasons totally unrelated to surface.
TFUS: And how do you think Kahneman is relevant to horseplayers?
JF: From Daniel Kahneman I learned to adopt an attitude like this one: Our minds are primitive. They play tricks on us. All day long we live by the absurd notion that “What you see is all there is.” We fail to account for the fact that we are actually seeing only a tiny fraction of what there is. We see a politician on TV and he seems smart and informed. We want to vote for him. But we do not see him all day long. We do not see that he is corrupt. Or we see a guy who dresses shabbily and eats in fast food restaurants. That is all we see of him. So our minds tell us that the guy has no money–when in fact he owns the fast food chain and likes to dress comfortably and couldn’t care less what some bozo busybody thinks about his clothes. What we see is NOT all there is. What we see is only a tiny fraction of all there is. That horse who just lost his debut by 30 lengths? That is all we see in the PPs, but the PPs do not display the fact that the horse had a toothache and refused to put out his best effort, and his connections are about to take advantage of that fact by cashing a big bet when the horse makes his second start. What we see is NOT all there is. Horseplayers who refuse to accept this are about to take a very long bath. As Shakespeare put it: “There are more things in heaven and earth, Horatio, than are dreamt of in your philosophy.”
TFUS: But there are differences between randomness, nothingness and ignorance of subtleties. Does successfully playing longshots hinge more on knowing subtleties or on exploiting the way the human brain plays tricks on us, the way the human brain fills in gaps to give us too much confidence etc?
JF: This is hard to answer because it’s hard to put specific weights on things. To me it’s a matter of bringing to the table everything you can, everything that you think has any value at all, and then hoping that through experience, your mind learns to present you with a semi-coherent picture of how best to apply it. In other words, what may seem like information overload to you in 1997 can begin to take some sort of coherent form by 1998, and by 1999, you have an approach that you are thoroughly wedded to. I think Andy Beyer did a beautiful job of explaining this process in his book “The Winning Horseplayer,” much of which was about how he added trip handicapping to his arsenal. And, while I don’t have time to follow New York racing closely, I love listening to Andy Serling talk about horse races. He just strikes me as someone who has found a way to strike a beautiful balance between the respect one needs for the complexity of the game, and the necessity of actually taking all this knowledge and presenting it in a coherent, articulate way.
TFUS: Talk about the prominence of trip handicapping for you. How important is it to you to watch a race to see how a horse made its way around the track? And how important are trainers to you?
JF: I come from a harness racing background as well as a Thoroughbred background. In harness racing, as you know, trips are king. I brought this attitude with me when playing Thoroughbreds, and to this day, far and away the most important factor to me when it comes to playing longshots is trips–meaning underrated performances or overrated performances. On the SoCal dark days every week, I sit down with the results charts. On these charts I write down all of the TimeformUS pace figures and speed figures for the race. Then I watch the tapes with the pace at the front of my mind, make notes on the horses and the races, and fill out endless stable mails. This process plays a role in perhaps 50% of the longshots that I end up playing. Trainers are another key thing. And, while I do look at trainer stats before every race, what I take most seriously when it comes to trainers is not stats per se, but rather what Dave Litfin, in one of his books, called “things you can sink your teeth into,” meaning specific patterns related to specific trainers–the types of things you won’t learn unless you play the races daily or sit down with a mountain of PPs and study these guys.
TFUS: We get surprised to see the promotion of trainer stats in isolation. Can you generally comment on single factors as drivers of bets vs. multiple factors. Our impression is that if the horse looks good for 1/2 dozen reasons, he’s going to be bet for all or most of them and he’ll never be a long price. But betting a horse for a single factor (such as a trainer who has been successful third off a layoff) seems too flimsy to us.
JF: Again, I am not going to knock somebody for doing something that he can make work. But for me, broad trainer stats and broad category stats are for background at best. People look at trainer stats and see that a guy wins at 20% off the layoff, or has a strong ROI off layoffs, or, as we capture it succinctly on TimeformUS PPs, gets a rating of 93 out of 100 off the layoff. Fine. The guy deserves his reputation as a “layoff trainer.” But the “Uncertainty Principle” is at work here. To measure something is to alter it. To measure (and publicize, or even bet based on) a trainer’s ROI off of layoffs is to lower it. And if you look at these stats, you’ll often see their potency decrease as time goes on. Moreover, returning to Kahneman, what you see is not all there is. Using layoffs as an example, a guy might get 20% winners off the layoff, but if you look at the actual races that the stat is composed of, and look at workout reports for the individual horses, you will see that the guy wins at 30% when his horses are training well and only 10% when his horses are training poorly. And you’ll see the same effect when you do other analyses. And soon you’ll find that the positive-ROI stats can be made to disappear with ridiculous ease. My mind rebels against all of this, and I think it’s a mistake to oversimplify it. I much prefer something more specific, something more “crunchy,” less well known, such as “Trainer X has a lousy win percentage off layoffs, over a big sample, but when his layoff horses have been training well, and they have a long workout in their pattern, and they return to the races in a reasonable spot, and the trainer uses one of his go-to riders, the horse has an excellent chance to outrun his odds.”
TFUS: What do you think of the popular line about how bettors should never bet against a “legitimate favorite”?
JF: That line was popularized by the 1980-1990s author Dick Mitchell. I read his books years ago and enjoyed them, but I totally disagree with him on this particular matter. Mitchell was a rare thing in that he was a mathematician who could actually write well. He was capable of phrasing things memorably, or he’d present them in a mock-Biblical manner. So now it’s 30 years later and we’ve got people running around saying “Thou shalt not bet against a legitimate favorite,” and people believe this without bothering to think it through. First of all, define “legitimate.” What does it mean? Class of the field, speed of the field, ready to produce his best today, as Mitchell has it? Fine. Now put some “fair odds” on this horse and go on about your business of assessing the longshots. There is absolutely nothing about the presence of this “legitimate” favorite that automatically precludes a longshot in the same race from being a good bet. The “Don’t bet against legitimate favorites” line is a perverse variation on double counting. Yes, the longshot will probably lose. That is what longshots almost always do: They lose. And yes, it is quite likely that he will lose to the legitimate favorite. But so what? If the longshot is an overlay, the wins, by definition, will make up for all the losses. This is true whether the favorite is “legitimate” or the bastard offspring of a Budweiser Clydesdale.
TFUS: Have you always been a longshot player?
JF: No. When I was just starting out, every bet I lost hurt, and every bet I won left me feeling overjoyed. With that mindset, naturally, I was looking to play the most likely winners. It was only later that I developed what I’ll call an immunity to loss-induced despair. And it’s a good thing I did, too, because when you play a lot of longshots, you are going to become very well acquainted with losing streaks. Without this “immunity,” I would probably have spent the last 15 years sitting in a chair, staring at a wall, and muttering about a photo finish that went the wrong way.
TFUS: Was it an accumulation of events that turned you into a longshot player, or was it one race in particular?
JF: Some of both. But I can say what was by far the most important single race. It was about 15 years ago. I was handicapping a race that was to be run the next day at Oak Tree or Hollywood. It was a maiden race. I was looking at a second-time starter who was a big price on the morning line. His only race had been at Pomona or Fairplex, whatever they were calling it back then. On the speed figures I was using, he was slower than the top contenders by several lengths, and I was thinking that second-out improvement might close the gap. Then I dug deeper into his speed figure, which had been earned on a wet track. I had been making my own speed figures from about age 12 and was always fascinated by the process. So I looked up this wet-track day and saw what the figure-maker had done with it. He had split the card in two for purposes of making the variant, which is not unusual on such days. What interested me was that he had put my maiden race on the “track is fast” side of the dividing line and all the subsequent races on the “track slowed down” side. In other words, he had taken a race full of developing, lightly raced horses and assumed that the track had slowed down right AFTER that race. That is when I became completely obsessed. What if the track had slowed down right BEFORE that race? Looking at the PPs of these lightly raced horses, I felt there was no way to be confident in either direction, which meant that there was a pretty good chance that this longshot I was interested in was a lot faster than the speed figures said he was. So I decided to make a do-or-die roll of the dice and basically bet my entire remaining horse racing bankroll (which wasn’t much, but still) on this horse. Well, as you have surely guessed by now, the horse won the race and paid something like $120.
TFUS: How did this impact you as a player?
JF: It wasn’t the money so much. I might have lost all the money back over the next month or two. I don’t really remember. What I do remember is the feeling I had as the horse rallied to win the race rather easily. It was as if the stretch run were unfolding in slow motion and all of the tension was running out of my body. But much more important was the following: I had put a tremendous amount of effort into that race. I spent hours on it. And I had come up with an opinion that practically nobody agreed with. And my horse won. And this link between the effort and the reward, the feeling I got from cashing that bet when the tote board was saying so unequivocally that I was wrong, it stays with me to this day. I love the hell out of the effort involved in handicapping horse races. I love it as I love practically nothing else. And what I learned about myself is that I get the most joy out of handicapping when I am disagreeing with as many people as possible. I’m fairly certain that this is a psychological thing. I probably have some malady or complex or something where I need to define myself in opposition to others. It’s probably related to why the political candidates I vote for are lucky to get 1% of the vote. But in any event, the joy that I get from this whole process very much outweighs the negative feelings I get from the long losing streaks that come from playing longshots.
TFUS: Any psychological advice for people who are considering building an approach that emphasizes longshots?
JF: If you think you might be cut out for it psychologically, try it. But don’t bother trying it with paper bets, meaning tracking theoretical results without betting anything at all. Paper bets are worthless when it is a human being who will be doing the real handicapping. Pick the smallest bet size that still has you caring about the results of the race. Then gradually build it up if the results justify doing so. Then be honest with yourself and try to determine whether the pleasure and the gain justify the pain and the misery and the feelings of despondency and self-loathing. Oh, and one more thing: Early on in the experiment, if you find yourself ready to bet a massive longshot, and then you change your mind and decide not to bet on him, hedge instead. Don’t leave yourself open to him winning and paying a fortune while you have nothing on him. Such an experience can sour you on the entire project. Instead, pick a bet size that will leave you not knowing whether you want the horse to win the race or lose it. That is the perfect size for a hedge bet.
TFUS: Let’s go back to the sort of handicapping logic that informs your On The Contrary column. Are there single factors you’ll bet on with great conviction or frequency?
JF: Yes. Single factors: Interesting trips would be number one (with the caveat that seemingly interesting trips in a turtle race are nowhere near as interesting as they seem). Speed figures if we have a dramatic disagreement with the figures of others and, as so often happens in these situations, the odds are there to compensate for the risk. Crunchy trainer patterns. Speed figure patterns that I love. A powerful likely pace scenario. Once in a great while, breeding that I love. Strong track biases, but only if almost everybody in cyberspace disagrees with me. (The best way to determine whether there is actually a biased racetrack is to read the bias declarations of people in cyberspace and believe the exact opposite of whatever they are screaming at the moment.)
TFUS: And are there other factors you’ll be far more likely to only bet on in combination with others?
JF: Factors that I will not play in isolation but will play in combination: Workouts number one. If you are about to play a horse solely off of a glowing workout report, don’t. It does not work. Well-done workout reports are valuable but they are not stand-alone factors. Hidden class drops, including ones that are taking place even though the horse is running at exactly the same race condition as last time. Etc. Etc. Etc.
TFUS: On a day-to-day basis you’ll write up the sort of handicapping logic that informs your selections. Could you give us some general comments on handicapping longshots here?
JF: I tend to put playable longshots into one of two categories:
A: Those who have ALREADY proven to me that they are better than the crowd thinks they are and seem primed to fire today. In other words, longshots who have what I think of as hidden fundamentals that they are about to put on full display for the first time. Often these horses are lightly raced.
B: Those who I think are ABOUT TO prove that they are better than the crowd thinks they are. These horses are more speculative, and the case for them often contains moving parts.
To me, type A above is very often a Grade 1 bet. Type B above is generally (but not always) a secondary bet. Here’s a list of general concepts I keep in mind when looking for viable longshots to bet:
1: Don’t look for efficiency in horses. Look for inefficiency. Inefficiency in running style. Inefficiency forced on a horse by his trip. Inefficiency forced on a horse by his rider. Inefficiency of energy distribution. Self-caused inefficiency due to rankness. Any kind of inefficiency. Inefficiency is more desirable than efficiency. Why? Because if you’re efficient, there is little room to become more efficient. If you’re inefficient, there is a lot of room to become more efficient. And to win, longshots almost always need to improve in some area.
2: When it comes to trip handicapping, the least important thing that happens before the top of the stretch is 10 times more important than the most important thing that happens in the stretch.
3: If you lack time for trip handicapping, watch only the head-on of the break from the gate.
4: Yes, horses who receive jockey switches from top riders to bad riders win at a lower rate. You will see this if you look at the statistics, and racetrack bettors are profoundly aware of this move and bet accordingly. However, certain trainers are also highly, highly conscious of how the crowd bets these rider switches, and on occasion, they will take advantage of this fact by using unpopular riders. So you end up with a poker game in which certain trainers will fold, fold, fold, and then suddenly stay in and lower the boom at high odds. If you see trainers such as Pete Miller or George Papaprodromou (just to pick two names out of a hat) making a negative rider switch, or using a no-name rider, don’t be too quick to assume the worst. It might prove better if you entertain the possibility that what you see is not all there is.
5: A racetrack puts the races in a certain order, but that doesn’t mean you should be handicapping the races in that order. Indeed, at most tracks, you would be much better off handicapping the races backwards. Attack the big fields when you are fresh, and look for reasons to blow off the first race altogether so you can have a look at the racetrack before you make your first bet.
6: Racetracks and horseplayers will tell you over and over that you should be interested in a particular stakes race, but that doesn’t mean you should be interested in betting on that particular stakes race. Really, it boggles my mind why bettors obsess over most stakes races. Do bettors get a share of the purse? Focus on the races that seem to provide the best opportunities for your style. If those races happen to be stakes races, fine.
7: If possible, start handicapping two or three days before the races will be run. The further away from the race they are, in terms of hours, the looser bettors tend to be. As the race approaches, many bettors tend to tighten up and become more chalky. Capitalize on the handicapping looseness you will feel 48 hours before the race goes off.
8: When playing longshots, forget about “keeping a positive attitude” and all that other New Age nonsense. Instead, you should actively expect all your plays to lose, and you should visualize how you will keep your equanimity when they do.
9: Don’t celebrate with lobster after good days. Go out for lobster after bad days. That’s when you need it. After good days, pick up a pizza and go straight home and start handicapping again. The loose feeling that follows good days is very fertile terrain for handicapping longshots.
10: Do not believe anything positive that people in this sport say about their private betting results. A bigger source of worthless babble has never been discovered by man. Anything that people tell you about their private betting results needs to be filed at the bottom of the Hudson River. Should you think about what they say about the subject of handicapping? Maybe. Should you instantly believe it? No. What really matters in this sport is what you learn, or confirm, or refute, through your own efforts. That is what will, as Bruce Springsteen put it in a song, “get your facts learned.” Everything else is just talk.
Related Post: New To Betting on Horses?
If you’re just looking for some crucial links on how to use TimeformUS PPs…