1. Welcome to SportsJournalists.com, a friendly forum for discussing all things sports and journalism.

    Your voice is missing! You will need to register for a free account to get access to the following site features:
    • Reply to discussions and create your own threads.
    • Access to private conversations with other members.
    • Fewer ads.

    We hope to see you as a part of our community soon!

A gambling thought experiment

Discussion in 'Anything goes' started by Dick Whitman, Apr 7, 2015.

  1. bigpern23

    bigpern23 Well-Known Member

    No question about it. Having the liquidity to withstand the losses is essential, because there will undoubtedly be some losses. I loved the 60 Minutes piece on Billy Walters a few years, wherein he speaks about having like $2 million or $3 million at risk on any given NFL Sunday. If he goes 4-12 with that much at risk, he's losing a ton of cash, but it's nowhere close to his entire bankroll.

    In fact, he has enough liquidity that he intentionally bets against what he thinks will happen to move the line and, once it has moved, he has his proxy bettors simultaneously place larger bets on the other, more favorable side so he can clean up. Walters is the kind of guy - or more realistically, syndicate - that Ragu is talking about.
     
  2. The Big Ragu

    The Big Ragu Moderator Staff Member

    Yeah. I know I keep bringing it back to trading, but it is SOOOO similar, not just in terms of your odds of competing based on fundamental research (or information), but in terms of money management. A rule of thumb, for example, that often gets recommended for beginning futures traders is that they don't risk more than 1 to 2 percent of their trading capital on any one trade -- because of those runs where you might have weeks where NOTHING is working. In the case of those futures, the leverage is so great that you can get wiped out easily if you are overlevered (and most don't realize they are).

    Let's say you somehow have shifted your probabilities by doing something, so that over time, you now can expect to be right 60 percent of the time? You'd expect to make money over time. But there is nothing guaranteeing that over a relatively short period of time you aren't going to come up wrong several times in a row. And if you get wiped out because of it, you are victim to having been right in big the picture, but ultimately wrong because you didn't consider the little picture.
     
  3. LongTimeListener

    LongTimeListener Well-Known Member

    I would also add that in college basketball in particular you could be dealing with a fairly high chance -- much higher than anyone would like to acknowledge -- of the game being fixed. It's so easy to shave points in a midweek low-interest game. Imperceptible. Miss two three-pointers on purpose, but hit the rim, and voila! Get paid.
     
  4. YankeeFan

    YankeeFan Well-Known Member

    A couple of my buddies did this for couple of years. I think they focused on the Missouri Valley Conference -- I seem to remember Creighton being one of the teams they followed.

    The first year, they did very well. They scoured all the news they could. Read campus newspapers. Focused on finding out about potential injuries and such. Watched every available game.

    But, in years two and three their luck ran out, and I believe they've dropped it by now.

    I think part of the problem is that they upped their bets, believing they had figured out the universe, and got beat on some big ones. And, that's usually how it happens. Think about any gambling experience. There are hands in black jack that you should win. So, you split aces and/or double down, but still end up getting beat.

    You can find a game with a line that is "off", but it doesn't mean you will win.

    You'd think that over time, if you find enough of those, you will come out ahead. But, it's really hard. There's the vig. There's the randomness of it. There's the discipline required to not bet too much on any one game. And, there's often not enough of these games to with lines that are "off", so you end up betting on more games than you should.
     
    Last edited: Apr 9, 2015
    sgreenwell and bigpern23 like this.
  5. Dick Whitman

    Dick Whitman Well-Known Member

    That's like the whole Martingale system thing - you would 100 percent, without question, win over time. So roulette bets are capped, I think. And, as well, not many people would have the money available to make the inevitable hundreds of thousands of dollar bet that's going to come up after a run of 20 blacks or reds in a row.
     
  6. doctorquant

    doctorquant Well-Known Member

    I just did a quick-and-dirty simulation using Excel and the standard sports-book odds (11/10). In this simulation, I assume that: 1) you have a 10 betting-unit bankroll (i.e., $10,000 with which to make $1,000 bets); 2) you find 20 games a season in which it is a certainty that you have a 55% chance to beat the line; and 3) if, in a given season, you run out of money, you don't get to bet any more (that season). I simulated 1,000 of such seasons.

    Under these conditions, which are VERY favorable to the "It should be doable," the average seasonal profit was around 10%.
     
  7. bigpern23

    bigpern23 Well-Known Member

    I think you realistically need a 50 or 100 betting-unit bankroll. You shouldn't really risk more than 1-2 percent of your bankroll on any one game. I'd be curious, if you can do this easily, what the return would be.

    Betting 10 percent of your bankroll on each game is wouldn't be advisable no matter your odds.
     
  8. The Big Ragu

    The Big Ragu Moderator Staff Member

    A Martingale model in gambling. ... is also known as "stochastics" in trading.

    That is more along the way of how I think and operate -- although I don't use stochastics (have never found a consistent edge there. ... and you wouldn't believe the lengths I have gone to to back test several stochastic indicators!). Speaking very generally, stochastics tries to predict outcomes in random environments. I'd suggest you are way more likely to have success thinking about gambling THAT WAY. ... than you are thinking you can do better analysis than everyone else. Although, it's obviously not a simple thing to do either, and one that isn't likely to be successful either.
     
  9. Dick Whitman

    Dick Whitman Well-Known Member

    Right. I've thougth about using Martingale for gambling. For example, take some NBA team for an entire season and Martingale them. What's the longest losing streak ATS a team has ever had? I'm curious about that.
     
  10. doctorquant

    doctorquant Well-Known Member

    I'm not surprised. The efficient markets hypothesis has its detractors, but it's pretty clear it stands up to scrutiny when considering the kind of thing you're talking about.
     
  11. doctorquant

    doctorquant Well-Known Member

    I was getting ready for class and didn't get to follow up and play around with my simulation. I realized in the interim that I was possibly giving an inflated value for the return, because I was comparing the bankroll at the start of the season with the bankroll at the end of the season. Now I've redone it so that it looks at the total amount you win relative to the total amount you've wagered.

    In my simulation, I'm still assuming you have 20 "locks" (i.e., where your chance of beating the spread is 55%) a season. Each season, you start over with a fresh bankroll of X betting units. If there's money in your bankroll, you bet. If not, you don't.

    With X = 10 betting units, across 10,000 "seasons" I've observed average profits of around 5% to 6% (it's surprising how much variability there is even in such a large sample).

    With X = 100 betting units, the average profit is about the same.

    Now, with X = 10 betting units, you go broke around 2% to 3% of the time. The thing is, going broke is rather neutral: It keeps you from winning some of it back, but it also keeps you from losing any more.

    Again, this is a very, very favorable treatment of how the "it can be done" scenario would unfold. Twenty 55-percenters (i.e., my chance of beating the spread is 55%) in the course of a season is, to my thinking, very, very optimistic.
     
  12. The Big Ragu

    The Big Ragu Moderator Staff Member

    I'd make people's eyes glaze over more than I typically do on here if I posted all that I could say about this! But. ...

    Have you ever been walking down the street and you saw a $20 on the sidewalk? EMH would tell you that the $20 shouldn't be there.

    But it is.

    EMH would tell you that Warren Buffett doesn't see unappreciated value. Or that Paul Tudor Jones didn't have an edge of some sort. They have just been incredibly lucky or they take on incredible risks (and have dodged bullets).

    Is that really the case?

    I believe in EMH in the general sense of how markets work, but not in any absolute sense. I believe that in that general way, something close to EMH probably explains why 99.9999 percent of us can't compete on fundamental research -- or information.

    At the same time, I watch markets every day, very closely, and even though I can't prove it, I believe that there are behavioral factors (think crowd psychology) that influence price movements.

    EMH would suggest that is nonsense.

    I know from actually having put A LOT of money at risk, that making money is not as easy as simply knowing that crowd psychology is at work. I am not discussing this theoretically. Even someone who just has backtested various things endlessly, and thinks they have discerned something, isn't the same as a person who has money at risk. It all changes when you are sweating because a market is moving against you.

    I believe (although maybe I am seeing what I want to see -- who knows?) I recognize that crowd psychology in action sometimes. ... but that said, I haven't found too many things that allow me to recognize and capitalize on anything in a systematic, non-discretionary way. Nor have I found a "system" or any way to easily make money.

    Which is why whenever I talk to people who are looking to get rich quick, or who have never put their money at risk, I try to save them from losing what they have. I know there is no holy grail.

    The only people I know who successfully make money at trading (I don't know as many gamblers) don't think about it as theoretically as I think you get to for a living. They just endlessly look for things that work more often than they don't work, and when they find something, they are incredibly disciplined in how they go about it. They're not as concerned with why it works, as the fact that it does work. And they are grinders, because the edges they believe they have feel miniscule. They tend to be defensive -- more worried about losing money than making money.

    Not trying to be crptic, but most of the things that I have found "work" for me -- and I am successful -- would be too difficult to explain on here -- not because it's too complicated for others or because I am so smart, but because it took years of experience and noodling for me to get to where I am. I'd sound like I am speaking a foreign language.

    What I can say, though, is that even though I can't explain why certain things work, I strongly suspect they relate to crowd psychology, and not just pure randomness. The things I have focused on (and that is because they work) make the most sense that way, at least.

    But at a certain point, I am not that concerned with why the "trend is your friend," (I know people who have been successful using various momentum-related methodologies) or why most markets follow certain rhythms that lend themselves to swing trading. I'm just concerned with the fact that those things ARE true and by having a discipline in how I went about it, I have consistently bucked what EMH would tell you.

    EMH would say I am lucky or taking way more risk than I realize (which is interesting in itself, because I am very focused on risk -- at least as I measure it by drawdowns).

    Maybe it's true. . ... Or maybe it's not. :)
     
Draft saved Draft deleted

Share This Page