Negative Expectation Misconception

[GameTheory]

I'm sure when you *do* start selling something, you'll become an authorized advertiser, right?

I am really just wondering what a "competitive connection" is -- not your whole algorithm -- just a hint?

[singunner]

Seeing how much I've gotten from this board and how much I've grown from it, it's not a question of whether I want to become an advertiser, but whether they'll have me.

As for "competitive connections"... they're.... the relationship between horses? I define everything in the program mathematically, explaining it in English is not my forte. Of course, the basis of handicapping is comparing horses, but there's comparing horses, and then there's comparing horses... Wow, that's not helpful. Let me put it this way: a sophisticated handicapper looks at maybe 5 or 10 variables and ends up with a few hundred/thousand data points per race. My program looks at 1 variable and routinely finds hundreds of thousands (sometimes millions) of data points. Where most people were looking for breadth, I looked for depth.

[Robert Fischer]

to the original point - absolutely correct. The slots are an easy reminder. It seems they are in a lot of places horseplayers play now, and some nice new places to play horses have been built around slots and casino-like games... Some of them proudly return what 80 - 90%?, but it is compounding..

on racing a positive expectation skill game- there is an attainable positive expectation somewhere up to about 50cents on every dollar. Somewhere, I sure don't know where... 50 is very high, and it would be hard to believe if someone is meeting or exceeding that, but it could be possible if you have a huge bankroll and can afford to be very selective and patient. 20-25cents on the compounding dollar is a more realistic master skill level. I am talking about consistently making your high value plays. There is a huge misconception that horseplaying is about a jackpot. It is a skill and a job. It requires skill and hard work. You can hit a pick 6 or a super high~5 or a huge superfecta... and say at the end of the year you made 50 cents on every dollar = that is not what I am talking about. Unless you made those plays with a positive expectation (you had a value edge on the pick 6 or super and then you only wagered the appropriate portion of your bankroll). Otherwise it is gambling. Nothing wrong with gambling winning or losing. It does most often cloud the discussion of positive expectation limits.

on take - take is the fee you gladly pay to play The Public in a parimutuel game. Before you invest any money you make a conservative estimate as to the hit% of the outcome and the payout for the hit. You make this estimate already accounting for the takeout. My horse has at least a 40% chance of winning and it will go off at least 2-1. Thats 20cents on the dollar after takeout. Play like this and allow for some bonehead human error and I am up 10cents on every dollar for the month. How did I do it?? Maybe the public is betting a weak favorite or they can't see the way the track is playing... whatever.. If it fits your demands you buy the ticket. But if they want to raise the take another 10%- then they can go to hell!

[badcompany]

Originally Posted by Cangamble

If the average takeout rate was 40% it still wouldn't be a negative expectation game by definition, but at what point does one drop the conventional definition?

I don't think you can arrive at a specific % where the game is unbeatable (except, of course 100% takeout). It's more a matter of whether the game is WORTH trying to beat. At 20% takeout, ~2% of the players would be able to show a longterm profit (Formula2002 used a standard dev equation in another thread that arrived at this conclusion) and the amount of time those 2% invest would make their hourly wage very low, because, even if you're very selective in the races you play, you still have to look at as many races as possible in order to spot opportunites.

The point I'm trying to make is that if you beat the game but make less than minimum wage doing so, did you really beat the game?

[singunner]

Originally Posted by badcompany

The point I'm trying to make is that if you beat the game but make less than minimum wage doing so, did you really beat the game?

There are very few games in the world where twenty out of every thousand can make a long term profit playing them. There are very few games in the world that even offer a way to get back the money you put into them.

[Cangamble]

Originally Posted by singunner

There are very few games in the world where twenty out of every thousand can make a long term profit playing them. There are very few games in the world that even offer a way to get back the money you put into them.

Betfair has an average 3% takeout, and that is only when you win.
That is on all types of bets including sports betting.
But even sports betting averages around 5% when you take winning and losing into consideration with bookies.
And the odds in sports betting aren't a 50-50 flip of the coin. They are made so that each side theoretically will have the same action on them, so you are really betting against the public.
I'm not impressed with the 20 out of 1000 winners, no do I think it is accurate. I believe it is less in todays environment without a rebate.

[jfdinneen]

As we are all aware, given the parimutuel nature of our sport, it is imperative to find overlays in the various markets in order to be profitable over the long-term.

Given this basic requirement, I would advise all beginners to calculate a simple "Actual / Expected Index" (AEI) at least monthly (if not weekly) in order to gauge how well they are performing against the competition. For example, if we look at win markets only and if we assume that the best estimate of a horse's probability of winning is the public odds ("Wisdom of Crowds"), then we can calculate the AEI as follows:

Expected = SUM(1/DecimalOdds) [For all bets]
Actual = COUNT(All Winners)
AEI = (Actual / Expected) - 1 [If positive, then you are currently finding overlays and you probably have an edge over the competition]

For other markets (e.g., exotics), the calculation of expected winners and actual winners is a little more complex.

John

[DeanT]

Originally Posted by badcompany

and the amount of time those 2% invest would make their hourly wage very low, because, even if you're very selective in the races you play, you still have to look at as many races as possible in order to spot opportunites.

The point I'm trying to make is that if you beat the game but make less than minimum wage doing so, did you really beat the game?

Just an honest question: Have you known a winner? I know a few, and they sure as hell aint working for minimum wage. If their bankroll was like $80 maybe, but not when you are slinging thru hundreds of thousands, or millions per annum.

You can make 0.5% ROI in this game and make significant money.

[badcompany]

Originally Posted by DeanT

Just an honest question: Have you known a winner? I know a few, and they sure as hell aint working for minimum wage. If their bankroll was like $80 maybe, but not when you are slinging thru hundreds of thousands, or millions per annum.

You can make 0.5% ROI in this game and make significant money.

I've never met a professional horseplayer, but, I'm sure there are a FEW out there. There was one in last year's World Series of Poker, his last name was Rosenthal. The commentator was making jokes about it, "No one wins at horses!"

That said, I live on the Upper Eastside of Manhattan. One could make a strong case that there are more wealthy people here than anywhere in the country. I'd bet not a single one would list Horseplaying as his primary source of income. If betting horses were such a great game, why wouldn't any of these masters of the universe have gotten in on it?

I stand by my opinion that horseracing is a great pastime, if not taken to excess, but is a horrible way to make money.

[ryesteve]

Originally Posted by badcompany

If betting horses were such a great game, why wouldn't any of these masters of the universe have gotten in on it?

Who said it was easy to make a living at it? Not only is it hard, it requires an unusual emotional makeup. I understand where you're coming from, since it's not something I would ever attempt, but at the same time, I can't close my eyes to the fact that there are people doing it, and doing quite well at it.

[Bill Olmsted]

Any seasoned horseplayer has the requsite skills to make serious money trading stock via a discount online broker. At $7.00 a trade, the "take" is zip when you are buying and selling thousands of shares. Best, you don't lose your entire bet if your stock or option goes against you; the advantages of stock trading over horseplaying are just too numerous to list. So if it's pure money you're after, forget about horses and learn how to trade. But horseplayers do possess an "unusual emotional makeup" so we gladiy accept this ultimate challenge.

B

[DeanT]

Originally Posted by badcompany

That said, I live on the Upper Eastside of Manhattan. One could make a strong case that there are more wealthy people here than anywhere in the country. I'd bet not a single one would list Horseplaying as his primary source of income. If betting horses were such a great game, why wouldn't any of these masters of the universe have gotten in on it?

How can one argue with such indefatigable logic

[Kelso]

[QUOTE=singunner]
As for "competitive connections"... they're.... the relationship between horses?
/QUOTE]

Is this similar to the BRIS method of rating horses based upon how they've finished against other horses ... and how those other horses raced aginst yet other horses ... etc, etc., etc?

The net result, in gross shorthand, is a ranking of all horses based strictly upon a circle of competitive finishes.

[highnote]

Is this similar to the BRIS method of rating horses based upon how they've finished against other horses ... and how those other horses raced aginst yet other horses ... etc, etc., etc?

The net result, in gross shorthand, is a ranking of all horses based strictly upon a circle of competitive finishes.

I didn't know BRIS had a method like that. Do you know what it's called and how it works?

[GameTheory]

Originally Posted by swetyejohn

I didn't know BRIS had a method like that. Do you know what it's called and how it works?

From the "library" section at Brisnet, class ratings faq:

WHAT IS THE THEORY UNDERLYING THE RACE & CLASS RATINGS?
The Race & Class Ratings are based on the premise that a horse's performance in any given race can be accurately gauged by considering the horses which the individual has beaten; those that have beaten him; and by the margins involved. Final Time is NOT a factor used in calculating the ratings.

HOW ARE THE RACE & CLASS RATINGS CALCULATED ?
Using proprietary computer algorithms, the recent Class Ratings for ALL horses in their recent races are calculated, adjusted and then recalculated every time ANY horse crosses a finish line. This calculation process creates complex inter-relationships among ALL thoroughbred racehorses regardless of the track, distance, surface, or country the horses actually ran at. What makes the Race & Class Ratings such a powerful handicapping tool is the daily updates and constant adjustments made to the ratings which reflect the subsequent performances of ALL horses in EVERY race. The net result of this extensive data crunching is reliable, objective data which can accurately quantify the relative merit of any thoroughbred racehorse's performance.