Teenage Fun at the Track
I grew up near a horse racing track.
As a teenager, my friends and I didn’t partake in normal teenager activities. As a bunch of geeks, we didn’t go to many school dances or parties.
Instead, we hung out in a smoke filled race track with a bunch of old men.
I instantly fell in love with the track. The atmosphere was something that I gravitated towards.
There was something fun about 20 screens with odds and potential bet combinations flowing at me.
There was the smell of smoke everywhere, along with constant yelling and cursing.
I loved it.
My friends and I would always laugh at the old men who could string together the most colorful combinations of profanity.
They would invent combinations of curse words that I couldn’t imagine. These men were the William Shakespeares of profanity. Maestros of expletives.
We were never carded, despite the fact that we were under 18. It was strange because we were clearly a bunch of 16 year olds. We would place our bets electronically and only go up to the tellers to cash out.
I’m sure that the tellers knew what was going on and just didn’t care. They probably weren’t being paid enough to make a big deal about it. After all, our age was even more apparent when compared to the group of 70 year old men that we were hanging out with.
We would usually go to the track with $5 in our pockets and spend an afternoon at the track, pooling our funds together to share in the losses and occasional winnings.
The amount of effort and brain power that we put into losing money was truly astounding, but we had a lot of fun doing it.
I learned a few early lessons at the track:
1) Horse betting is a game where the house takes 20% of the pool and is a sure fire way to lose money.
2) The worst way to gamble is to bet on the horse with the best odds.
3) Horse racing is a fun way to spend $20 on an afternoon of exhilarating fun.
Point #2 is the most relevant when it comes to investing.
When we first starting gambling, we would bet on the favorites and thought that made the most sense.
As we lost money betting on favorites, we realized that the favorite was nearly always over-valued.
It dawned on me that horse racing wasn’t about predicting the horse most likely to win. It was about finding a horse that was undervalued. It was about finding where the payoff would exceed the true odds of the horse.
It didn’t take sophisticated analysis to realize that the favorite was overvalued. This was obvious when analyzing the data.
To find data, we needed an important source: The Daily Racing Form.
Unfortunately, the Racing Form was an expensive document and we didn’t want to dip into our gambling pool to pay for it. We had to get creative.
Usually, to conserve our betting pool, we would sit patiently and listen for a particularly foul mouthed gambler. Usually, the tempo of the profanity was tied closely to the amount of money that they were losing. Eventually, this gambler would throw their racing program on the floor in disgust and vow never to return to the track. They would scream that the place should be burned to the ground, as if it were the city of Carthage and they were the Roman Empire. (Narrator: They would usually return on the following weekend.)
I would then take the mangled program off of the floor and use it to make betting decisions for the rest of the afternoon.
In the program, the best horse was usually given something like 2-1 odds by a professional handicapper. As the bets poured in, the horse was usually assigned odds of something like 1-1 because everyone wants to bet on the winner. In other words, the favorite typically became overvalued.
The favorites were guaranteed to pay out less than they were “worth,” a phenomenon made particularly punishing by the house’s takeout.
In other words, everyone knows that the favorite is the likely winner and the favorite typically became over-valued. Betting on favorites is a way to win more often, but it’s also the way to lose the most money at the track.
I think of the best stocks in the stock market as favorites at the track.
Charlie Munger explains this better than I can:
The model I like—to sort of simplify the notion of what goes on in a market for common stocks—is the pari-mutuel system at the racetrack. If you stop to think about it, a pari-mutuel system is a market. Everybody goes there and bets and the odds change based on what’s bet. That’s what happens in the stock market.
Any damn fool can see that a horse carrying a light weight with a wonderful win rate and a good post position etc., etc. is way more likely to win than a horse with a terrible record and extra weight and so on and so on. But if you look at the odds, the bad horse pays 100 to 1, whereas the good horse pays 3 to 2. Then it’s not clear which is statistically the best bet using the mathematics of Fermat and Pascal. The prices have changed in such a way that it’s very hard to beat the system.
Everyone knows who the winners are in the market. Facebook, Amazon, Google, Netflix, Visa, Mastercard, Costco, Domino’s, etc.
Everyone wants to bet on the winner.
As a result, the favorites are bid up to extraordinary valuations. Valuations that don’t always make sense.
It happens again and again in every market cycle. The Nifty 50. The late ’90s Nasdaq.
Everyone wants to bet on the winner. It’s satisfying because the winners win the most often. There is also comfort in going with the crowd.
However, betting on the favorites is a way to lose money at the track. Similarly, betting on expensive stocks is a way to lose money in the markets, even though it hasn’t seemed that way for the last decade.
Like the track, the trick in investing is to identify under-valued bets. You want to find a bet where the odds don’t make sense. You want to identify situations where the quoted odds are worse than the true odds. In other words, you want to identify mis-priced securities.
With that said, reflexively betting on horses with bad odds is also a way to lose money, as my friends and I learned. They often deserve those bad odds. This is likely why buying all of the stocks at 52-week lows is also a terrible way to invest.
Anyway, like favorites, expensive stocks under-perform over the long run as a group, which is what all of the data screams in every backtest over every long period of time.
Of course, it’s hard to rely on that data when looking at a market. It feels good to bet on the winners.
The trouble is: everyone else knows who the likely winner is. That’s why they command a high price.
The OG Of Quant: Andrew Beyer
In my quest to find a better way to handicap horse races, I looked for a book that would teach me how to handicap.
I discovered the work of the greatest handicapper of all time: Andrew Beyer.
Andrew Beyer was doing quantitative analysis before Moneyball. He was doing it before James O’Shaughnessy wrote What Works on Wall Street or Jim Simons launched Renaissance Technologies.
When Andrew Beyer started betting, most horse handicappers were obsessed with class. Class was based on the breeding history of the horse. Horses of a better lineage were assigned better odds. Handicappers would often assign nonsensical stories to horses in their efforts to attempt to predict horse races.
Andrew Beyer thought that all of this was unreliable sorcery and sought to create a more rigorous way of determining the true odds of the horse.
He developed what became known as the Beyer Speed Figure: a quantitative way to determine how fast a horse ran a race and to do it in a way that captured different conditions and lengths of track. A single number that would assess a horse’s performance in a race.
When Beyer was gambling with his own funds in the 1960’s and early 1970’s, he had to manually calculate speed figures. He would spend his time manually going through old racing forms and figuring out the speed figures of different horses.
At the time, he had a critical piece of information that no other horse player had. He was like the Jim Simons of the racetrack.
Beyer was able to use this information to figure out that the true favorite in the race was a long shot with 30-1 odds that no one was paying attention to.
He had the holy grail of horse racing: a quantitative system that would allow him to collect money from other gamblers at the track.
In 1975, Andrew Beyer wrote a book about his discovery, Picking Winners. This began the gradual unraveling of speed figures as a tool to make money.
Andrew Beyer could have silently continued cashing in wining tickets, but he was a journalist and a young man who wanted to make his mark on the world, so he published his book about it.
The book outlined how to manually calculate speed figures, which many horse players started doing. Unfortunately, because more people knew about the signal, horses with the best speed figures started to command better odds.
In the 1980’s, paid services began providing the speed figures for a fee. This further diminished the labor of horse geeks who were manually calculating speed figures on their own.
Beyer responded to this by publishing The Winning Horseplayer in 1983, which advocated using complex bets (such as Exactas, Trifectas, and Daily Doubles) and combining with speed figure analysis to identify under-valued situations at the track that were not simple win bets.
Of course, the knowledge and availability of speed figures continued to chip away at the potential profits that could be made from this signal. Even the complex bets advocated in The Winning Horseplayer lost their effectiveness.
The final nail in the coffin was applied when The Daily Racing Form began publishing speed figures in 1992. Once that happened, the horse with the best speed figure started to command the best odds. Speed figures were no longer a tool that could be used to regularly generate profits.
With that said, speed figures were still the best tool available to predict the outcome of a horse race but they were no longer a way to consistently cash winning tickets.
Andrew Beyer became a legend and I’m sure he made a significant amount of money from the publication of his books and the incorporation of his method in the Daily Racing Form.
Unfortunately, Beyer speed figures no longer offered an edge to the horse player who could use them to find under-valued bets at the track.
Once everyone knew about them and the information was widely available, the party was over.
I sometimes worry that simple metrics of value are suffering the same fate as the Beyer Speed Figure.
Beyer himself discusses his speed figure in comparison to the stock market and the use of the P/E ratio:
It was a tremendous edge to have the figures at a time when most people didn’t use them or even believe in them. I can only draw an analogy to the stock market if the concept of the P/E ratio were unknown to, or its importance was disbelieved by the majority of people buying stocks, and you were about the only guy who knew what the P/E of different stocks was, it would be a tremendous advantage and I had that advantage for many years.
When everyone realizes that simple measures of cheapness work, doesn’t that crowd out the trade and reduce the effectiveness?
One can imagine that men like Benjamin Graham and Walter Schloss were like Andrew Beyer at the track in the 1960’s. They had a statistical method that no one else used and were able to use the information to make tremendous amounts of money in the stock market.
The disappearance of net-net’s in the US stock market lends credence to this view. Net-net’s disappeared because everyone knew about them. It’s obvious that a company selling below liquidation value is a tremendous bargain, which is why there aren’t a lot of companies that trade below liquidation value today.
Meanwhile, value as a statistical factor has become very popular since Fama and French made it respectable in the early 1990’s.
If everyone knows that simple statistical measures of value work, will it stop working?
Perhaps the “cheap” buckets of the market were filled with mis-priced securities in the past, but now this cheap universe generates more scrutiny among investors, thus eliminating the mis-pricings.
As the mis-pricings are eliminated and quantitative investors elevate the valuations of the cheapest stocks, will this have the same effect of everyone betting on speed figures and making the signal worthless?
I don’t think so.
If this were happening, I would expect to see the cheap buckets of the market become expensive relative to their long term mean.
This doesn’t seem to be happening. In fact, the cheapest chunk of the market is not expensive relative to its long-term mean.
If value were becoming a crowded trade, wouldn’t this segment of the market become more expensive? Just like horses with high speed figures started to command better odds as the figures became more popular?
I actually think this case could have been made in the mid-2000’s, but no one was making it back then.
By the mid-2000’s, value was triumphant. The unwinding of the tech bubble wrecked the investment world, and the only people with respectable track records were those who embraced cheap stocks.
Quantitative value investing became popular and investors piled into it. This made value stocks more expensive relative to their long-term mean.
We’re now at the opposite end of the spectrum. After a poor decade for value investing, most people hate it. Even value investors are capitulating and buying into the compounders.
I think the absolute cheapness of value is an indication that the party is not over for quantitative value investors. It’s as if the high Beyer speed figure horses are commanding worse odds than they were before the signal was discovered. It’s not a crowded trade right now.
It was a crowded circa 2007.
It should have been obvious in the mid-2000’s that this was becoming a crowded trade when value stocks started to become much more expensive relative to their long term mean.
For this reason, I think it is important for quantitative investors to play close attention to the absolute valuation of the value segment of the market. There are times when the bet on value makes sense and times when it does not.
It’s completely possible that simple metrics of cheap will no longer be useful in the future. Monitoring the absolute cheapness of the factor is a way to monitor whether or not this is happening, just like better odds on fast Beyer horses was a sign that the party was over.
The Equity Risk Premium
Everyone likes to pick on value investors, but I find it amusing that no one ever asks if something like the equity risk premium can be arbitraged away.
In my view, the equity risk premium is not different than the value factor.
If everyone knows that equities command higher returns than other asset classes, why doesn’t that become a crowded trade that reduces long-term returns?
So far, the 21st century hasn’t offered much of an equity risk premium. Since 2000, the US stock market has delivered a 5.62% CAGR, while the US bond market has delivered a 4.98% rate of return. That’s not much of a premium, especially compared to the pain that equity investors had to endure, such as multiple 20-30% meltdowns and two nasty 50% drawdowns. International stocks have actually under-performed bonds entirely, delivering a pathetic 2.58% CAGR.
This all stands in sharp contrast to what happened last century, before everyone knew about the equity risk premium.
In the 20th century, stocks returned 10.16% and bonds returned 4.82%. Now that’s a premium!
Could the poor performance of equities have occurred because everyone realized that equities perform better over the long run in comparison to bonds? As a result, were equities bid up to a level in the late ’90s that guaranteed poor performance going forward?
In fact, right now, equities are so expensive that they are likely to deliver another lost decade like the 2000’s. It’s possible that they may even under-perform the low yields offered by US treasuries. In fact, I think that’s probable.
At the moment, I think that the prospects for the value factor are better than those for the equity risk premium itself. This is high level heresy, but the evidence backs it up. After all, in the 21st century, there hasn’t been an equity risk premium. Small value is one of the only investing styles that has offered a decent return over the last twenty years, a 7.9% CAGR.
Two TV themes that struck terror in my heart as kid in the 1980s:
PLEASE NOTE: The information provided on this site is not financial advice and it is for informational and discussion purposes only. Do your own homework. Full disclosure: my current holdings. Read the full disclaimer.