[magyarsvensk]

I am not sure how to introduce this subject except to say that I will be glossing over a lot -- partially because I want to keep my work somewhat private and also because an in-depth discussion would get deathly boring.


In college, I was fascinated by the potential to apply artificial intelligence algorithms to the stock market and made it one of my senior projects. The first incarnation was an expert system based on one of Warren Buffet's books. This ended up being nothing special since one could easily get better results with a pen and paper and Yahoo Finance's stock search tool. But you know, you gotta start somewhere.


The biggest problem with using expert systems is that the subject of finance is too broadly studied and profitable. Expert systems are geared towards esoteric subjects that few know anything about and even fewer would want to learn.


I spent several years believing that there was no advantage to be had in the markets until I read the book Trend Following by Michael Covel a few years ago. More interesting than the interviews with the experts to me was just the basic concept: that trends have always existed and will always exist as long as humans are involved. The efficient market hypothesis is wrong or else how would these guys have made money? Furthermore, the Incompleteness Theorem would imply (to me at least) that the efficient market hypothesis cannot be true, but I digress....


Since then I have developed an evolutionary algorithm based hedge fund program that does quite well indeed....live simulations of my latest version have yielded a 17.6% return since April 15. The problem there is that the initial investment would ideally be very high and the potential returns are quite low.


That is when compared with the potential that the same formula could be used to develop a sports betting algorithm, which I have also been developing over the last two years. So far I have only done well in college football during the 2013 season, but there have been a lot of improvements made since then. The current version is being applied to baseball betting, and has done quite well indeed. I have broken even over thousands of bets so far this season as improvements have been made and bugs fixed, which would indicate a slight edge over random. The latest version is running at a very good clip though.


So has anyone else pursued the idea of a game-playing/trend-following algorithm for sports betting? Results good or bad? Pitfalls? Sticking points?


Now for something hopefully less boring....today's picks:


Toronto, Boston, Oakland, Chicago White Sox, Kansas City & UNDER, LA Angels, Baltimore, St. Louis, Arizona & UNDER, Washington, Atlanta, NY Mets, Pittsburgh, LA Dodgers, Miami & UNDER

[basket33]

lots of people are doing it, there is a lot data in sports now, every move players make is tracked on major sports and leagues... check http://www.sloansportsconference.com/
problem with sports after you find a system is that there is no real market yet, exchanges are still week and with high commission, it hard to place big bets ie. placing bets cost you to much, there are some improvements on asian market but its still like 30yrs behind stock market

[magyarsvensk]

I should clarify that this isn't sabermetrics at all -- it's not an academic study of a specific pattern, it's not a theory, it has nothing to do with "big data" (although I suppose it could) or even "statistics" really.

This is applying the same algorithms that are used for games like chess and bridge to sports betting. In other words, I am not trying to win by using data generated by the computer, the computer itself is trying to win.

[statnerds]

I spent several years believing that there was no advantage to be had in the markets until I read the book Trend Following by Michael Covel a few years ago. More interesting than the interviews with the experts to me was just the basic concept: that trends have always existed and will always exist as long as humans are involved. The efficient market hypothesis is wrong or else how would these guys have made money? Furthermore, the Incompleteness Theorem would imply (to me at least) that the efficient market hypothesis cannot be true, but I digress....

I would make the case here that you are confusing a handful of guys making money as evidence that EMT is wrong while ignoring the thousands on the other side that have lost money. Not trying to discredit or attack you or anything of that nature, but these 'guys' that made money are the fortunate winners of random results. Until you know how many other 'experts' lost doing the same thing, you lack enough evidence to tear down EMT. And we are only hearing of the success stories.

Bill Miller and whatnot.

Got 6,000 experts doing the same thing, but we hear about the 5 that mad money?

[magyarsvensk]

No offense taken. I enjoy a spirited discussion on the topic.

If 6,000 experts were doing the same thing, then they would have had the same results over long periods of time, right?

I hadn't heard of Bill Miller, but he seems like more of a buy-and-hold type investor rather than a trader. Trading more closely resembles sports betting. I am thinking more like Edward Sekoyta, Ed Thorp and other traders who don't make any decisions -- they just wake up in the morning, do what the computer tells them to do and take the rest of the day off. Those guys have achieved unreal returns.

Like any theory, the EMT is only as strong as the data that most (if not nearly all) market participants are using. So yes, if you use the exact same information and analytical tools as the next person, then you will achieve the same results over the long haul. If you use different information and/or different analytical tools, you will achieve different results over the long haul. The goal is to find the best information and the best analytical tools.

Early chess programs analyzed the board much in the way that humans do -- looking at diagonals, board control, things of that nature. They all failed miserably. It was only when a simple brute-force goal-oriented approach was applied that computer chess programs surpassed human abilities.

From what I have seen, the advancements that have been made in game-playing algorithms have not yet been made in investing or in sports betting -- perhaps because of the egos involved. It is much easier for your average asshole trader to say, "make a program that does exactly what I do except faster." Much harder to say, "I have no idea what I am doing, a computer could do it much better. I'm going to let the computer develop its own theory and apply it."

[Stefan Penner]

That's exactly right. And let's not forget the most fundamental problem that overshadows everything here:

If you actually have any controlled method which earns marked profit in the stock market or in sports betting, you will have every single motivation to let NOBODY know about it!!

You would, on the other hand, have every motivation to write a book on "the 5 magic stones that make me win" and then sell the book and the stones to poor suckers that see your success and are naive enough to think that you'll share the method with the general public.

In the stock market and especially in sports betting, if a sure fire method of positive returns was ever made public, all books would be forced to inflate the juice to compensate.

When the majority of bettors become better at it than the books, the books are forced to increase the vigor.

It's that simple.

Anyone that earns sports betting will also have the where-with-all to not tell anyone how they're doing it.

I'm fluxing between +200% to +400% my initial bankroll last 600 days.

You might grab some tips from me because I'm peanuts compared to the best that do this...

...but to the monsters in this game, don't expect them to bring you in. Expect them to attribute their success to everything BUT what they're actually doing to win, and expect to pay for the 'advice'.

[magyarsvensk]

Sure, yeah, that's all good and fine.

But isn't the purpose of this forum to discuss handicapping techniques? You could easily copy and paste your post into any of these threads.

[bihon]

So has anyone else pursued the idea of a game-playing/trend-following algorithm for sports betting?

I came here too from the financial trading world. Much less hussle for a small investor/gambler, less people involved, less juice, less paperwork... while the underlying industry is exactly the same: gambling (except for very few hedging constructions by financials). The only drawback in betting is few reliable bookies.

In both worlds the same strategies could be used, fundamental or technical analysis and for those going technical it is a perfect world as the order execution is instant and sure.
Now, going back to trending question I have to say that I support efficient market view as I made a lot of similar research in both worlds, constantly getting perfectly balanced results, of course in markets with high volume.
In short and simply put, the sum of all streaks is equal to sum of single (one up/down) outcomes.
If you can find a saturated market where this ratio goes to any side, you'll have the winner.

[statnerds]

I hadn't heard of Bill Miller, but he seems like more of a buy-and-hold type investor rather than a trader. Trading more closely resembles sports betting. I am thinking more like Edward Sekoyta, Ed Thorp and other traders who don't make any decisions -- they just wake up in the morning, do what the computer tells them to do and take the rest of the day off. Those guys have achieved unreal returns.

Like any theory, the EMT is only as strong as the data that most (if not nearly all) market participants are using. So yes, if you use the exact same information and analytical tools as the next person, then you will achieve the same results over the long haul. If you use different information and/or different analytical tools, you will achieve different results over the long haul. The goal is to find the best information and the best analytical tools.


From what I have seen, the advancements that have been made in game-playing algorithms have not yet been made in investing or in sports betting -- perhaps because of the egos involved.

Last thing first, just for reference:

along the lines Al Go Rhythms.

First part last now.

love Thorp. seems he never gets credit due, only in my mind, along with claude shannon. anyhoo, thrilled you took no offense.

agreed on early points and will now reveal a huge chunk of the basis of my approach to sports betting. they are stocks or any other asset than can be traded. books, the best ones, understand their role as a provider of a market place and nothing more. they charge their fees to provide said market place, but that should be expected/accepted. building on that i bet baseball more like trading (which certain secrets will remain so), but i also strongly believe the market is efficient as the punishment for being wrong can be financially severe. unlike Investment Managers, whom take similar risks, but with other peoples' money and with no appreciable difference in ability or results that cannot be explained away by randomness. this is getting too long...summary

baseball teams are stocks
book provide market place
market is efficient (my opinion, strongly)
beating the Market to the closer (positive CLV) will produce profits long term
And the final piece stays under my hat until i write a book revealing it.

i do like your starting point, maybe i just need to reread your posts and get a deeper understanding.

but trends (which may not fit exactly what you reference) made me think the most recognizable one for me is strong money on boston. everyday. armed with that information one could either bet them early and look for Arb later, or ride it and hope long term clv produces, or sit out and wait until last minute to fade sox and get best number on opponent possible.

[statnerds]

If you actually have any controlled method which earns marked profit in the stock market or in sports betting, you will have every single motivation to let NOBODY know about it!!

You would, on the other hand, have every motivation to write a book on "the 5 magic stones that make me win" and then sell the book and the stones to poor suckers that see your success and are naive enough to think that you'll share the method with the general public.

When the majority of bettors become better at it than the books, the books are forced to increase the vigor.

I'm fluxing between +200% to +400% my initial bankroll last 600 days.

You might grab some tips from me because I'm peanuts compared to the best that do this...

if truly successful, regardless of the method, the fundamental truth is you will run into limits. and selling the information carries the potential for much greater return with much less risk.

the books are not forced to do anything when a majority of bettors become better than themselves except make tighter lines, which goes back to EMT and the understanding that markets evolve constantly. so if this amazing method of winning was out there, one way or another, it would be incorporated into the market, increase its efficiency and render the method no longer viable. these things are not static.

as for 200% or more in less than 2 years, i would much rather hear that expressed as ROI as most professionals i have encountered would be thrilled with a return in the area of 3%. not saying it is impossible, simply stating that to triple your BR in 600 days, given standard rates of return from this sports betting thing you would be having nearly your entire BR in play daily which leads us to concepts such as Gamblers Ruin and whatnot. without running exact equations, and off top of my head, you would actually need 5X your BR in play everyday to hit that 200% in under 2 years. not saying it could not happen, just saying, cash out now.

[magyarsvensk]

Lots of good stuff in this thread now. Looks like the finance guys come alive on the weekends.

My job is finance related but mostly removed from the markets. I understand why people want to believe in efficient markets, but to me the concept sounds just as implausible now as it did when I first heard it in my Math of Finance class.

For one thing -- a big thing in my mind -- the efficient market hypothesis seems to contradict the incompleteness theorem. For all systems (including the markets) there will be an infinite number of true statements (trends, betting systems, etc.) that cannot be derived from the axiomatic foundations of a system.

In other words, if the market makers are using system A, system B, system C and system D to create an efficient market, you will always be able to find a system E that makes money despite the fact that the market has adjusted for systems A through D. And when and if system E gets incorporated, there will then be a system F that also makes money.

[Plaza23]

baseball teams are stocks
book provide market place
market is efficient (my opinion, strongly)
beating the Market to the closer (positive CLV) will produce profits long term
And the final piece stays under my hat until i write a book revealing it.

.

The main problem with that (in comparing sports gambling to investing in the stock market), is that to make money at gambling you do have to beat the market (since there is a winner and loser in nearly every bet).

Stock market is different. Based on historical returns, one would only need to BE the market, and they'd be making money. And its pretty easy to be the market in investing. You can buy into mutual funds, the SP500, etc.

I do think the stock market and gambling markets are equally efficient. Line movements are very similar to price movements. But, there's juice in gambling. Sure, there are fees associated with investing, but not to the extent of the juice in gambling. Then you also have the winning/losing dynamic. In sports, there's always a winner/loser. But investing isnt like that. Just because the stock price for IBM went up by a certain %, does not mean that another stock went down. That dynamic is not there in the stock market. Its not an even sum game. You dont have to be on the "winning" side to make money because there's not sides to the stock market. There's peaks and valleys and the wave brings everyone up and everyone down.

If someone wanted to devote a great deal of time to researching ways to make money - and they had either the stock market or sports gambling to choose from. I'd tell them to go play stocks. Just my opinion.

[magyarsvensk]

Yes, there is juice in sports betting, but the potential returns are much greater. Even with triple-leveraged ETFs, the daily swings will be on the order of 2%-3% most days, and a good average daily return would be on the order of 0.50% or less.

In sports betting, if you can consistently beat the vig, you are probably doing better than 0.50%. Your best betting day would probably net a return of 20-30%, whereas in investing, that is your best 1 or 2 month period.

The biggest advantages of trading in the markets versus sports betting are (1) a huge market that you could never move if you tried (thus guaranteeing if you have good system that it will always work), (2) no vig and (3) it's legal.

[thom321]

I would agree with Plaza that the stock market is less efficient than the sports betting markets. Valuations of stocks as a whole (and more frequently for individual stocks) repeatedly get far away from what an objective fair value would be due to human emotions etc. The same can not be said about odds offered by bookmakers. For anyone owning TSLA, FB, TWTR or even AMZN for the long run, please get back to me in 10 years and tell me how well that worked out.

While sports betting inherently has more built in leverage than stocks due to the margin requirement for stocks, other financial markets like forex offer comparable leverage and return potential. However, the forex market is also considerably more efficient that the stock market.

Pretty much all financial markets also have the equivalent of vig through the bid-ask spread plus commission and while it is generally less than the vig offered by even the best books, small cap stocks and the options market frequently have prohibitively high spreads (vig).

I do agree that short term trading (of financial instruments) and sports betting have a lot of similarities when it comes to proper risk management.

Yes, there is juice in sports betting, but the potential returns are much greater. Even with triple-leveraged ETFs, the daily swings will be on the order of 2%-3% most days, and a good average daily return would be on the order of 0.50% or less.

In sports betting, if you can consistently beat the vig, you are probably doing better than 0.50%. Your best betting day would probably net a return of 20-30%, whereas in investing, that is your best 1 or 2 month period.

The biggest advantages of trading in the markets versus sports betting are (1) a huge market that you could never move if you tried (thus guaranteeing if you have good system that it will always work), (2) no vig and (3) it's legal.