money management program or software

[buffaloxp]

Looking money management program or software for horse betting and sport betting, any input
thank

[Robert Goren]

A link for a Kelly Calculator

http://www.albionresearch.com/kelly/default.php
A warning about Kelly. If you over estimate your win percentage , you will die. So be very conservative with that number.

[offtrack]

Take a look at HMI software from Dave S at TheHorseHandicappingAuthority.

Good for my win betting, and I think it would be interesting to use with sports betting too.

[pktruckdriver]

No offense meant here, but my best Money Management trick was this ...

Keep your money,

Do not play seriously, just have fun with it, enjoy it once in awhile, on the weekend, or when you visit a track, but to do this seriously enough to have a mangement tool , then Offtrack may be right, use that one, it works, but keep my opinion in mind, okay.

Best of luck


Patrick

[lansdale]

Originally Posted by buffaloxp

Looking money management program or software for horse betting and sport betting, any input
thank

Hi Buffalo,

I assume from this question, that you're playing with an edge. If not, no money managment tool can help you. If you are, Kelly is, by far the most profitable money-management method, but also, like a power tool used incorrectly, potentially dangerous. It's especially difficult to adapt to horseracing, because, unlike some casino games, such as blackjack, it's much more difficult to form an accurate win probability assessment for every horse in every race you intend to bet. This is an absolute requirement. Horseplayers who claim to bet Kelly based on a rough estimate of win probability aren't betting Kelly. I believe that only database players with a thoroughly tested method are capable of using this kind of money management.

Cheers,

lansdale

[Lefty]

Huey Mahl used to suggest to recalculate everything every 15 races to use kelly on horse racing.

[lansdale]

Originally Posted by Lefty

Huey Mahl used to suggest to recalculate everything every 15 races to use kelly on horse racing.

Hi Lefty,

No disrespect to Huey Mahl, but this is a really bad idea. You can't uses averages as a basis to bet Kelly. And as I said, if you go wrong with Kelly the losses can be huge. In order to use Kelly, you really must know the math and you really must know your per race edge with great accuracy. Most horseplayers don't know either.

Cheers,

lansdale

[Lefty]

Maybe I stated it wrong, but you do the math but you redo it every 15 races to stay on top of things. I had a software program once that stayed in the black
for quite a spell and doing it Huey's way resulted in a big profit.
Huey at on time was a mathematician for Nassa.

[lansdale]

Originally Posted by Lefty

Maybe I stated it wrong, but you do the math but you redo it every 15 races to stay on top of things. I had a software program once that stayed in the black
for quite a spell and doing it Huey's way resulted in a big profit.
Huey at on time was a mathematician for Nassa.

Hi Lefty,

There's no question that Kelly works well. A number of people have made a lot of money using it. I use it to play blackjack. I do know that it's a much more difficult proposition to adapt for horseracing.

Re your quote, I'm not sure what it means. If you have a stable model, it shouldn't need to be updated twice a month. You say that you bet Kelly using a model and did well. If your model produced a per race edge and you were betting Kelly, it should still be making money. In fact, anyone who does have a long-term edge and bets Kelly is on their way to being a millionaire. My guess is that your model was flawed, hence Kelly stopped working.

I know that there are people on this site like Formula 2002, who use Kelly, and have the background to use it and not get hurt. I believe sjk is someone else who may use Kelly, and understands how to use it. A few years ago Dave Schwartz said Kelly was too dangerous as a betting tool. But after he improved his model, Kelly suddenly started working for him. This is really my point. If you have a model accurate enough to produce a consistent per race edge, than Kelly is for you. If not, it can cost you a great deal of money.

BTW, I'm familiar with Huey's reputation and background, but with all due respect, the people I've learned about Kelly from are far more qualified.

Cheers,

lansdale

[Lefty]

Huey did more work with Kelly than anyone I ever met or read about.
It's no more dangerous than flat betting. You set a bankroll and don't go
crazy.

Horse racing percentages fluctuate, that's why Huey updated it regularly, so as to make the necessary adjustments. I think that's the way not to get hurt instead of having a model you're hoping will stay stable.

My model didn't stay stable and that's why I bailed out. Precisely because I did update the model every 15 races is why I could still bail with most of the money.
Edges change constantly and very likely there will come a time where they will
disappear altogether. I think to think you will have a constant non-changing edge is insanity.

[Dave Schwartz]

... yet, without having a "constant edge," how do you know what your edge is?

[thaskalos]

Should the horseplayer perceive himself as having a certain, verifiable edge...the same as the casino does, in the games that it spreads?

Speaking for myself, I would be quite reluctant to forwardly project what my edge should be...because of the shifting ground beneath my feet.

Changes occur pretty rapidly in our game...

It's one thing to estimate your edge by reflecting backwards on your gambling activity...but to actually base your future wagers on the assumption that this edge will remain constant in the future?

That's a little too risky for me...

[lansdale]

Originally Posted by thaskalos

Should the horseplayer perceive himself as having a certain, verifiable edge...the same as the casino does, in the games that it spreads?

Speaking for myself, I would be quite reluctant to forwardly project what my edge should be...because of the shifting ground beneath my feet.

Changes occur pretty rapidly in our game...

It's one thing to estimate your edge by reflecting backwards on your gambling activity...but to actually base your future wagers on the assumption that this edge will remain constant in the future?

That's a little too risky for me...

Hi Thaskolos,

I'm assuming from your many posts that you're either a pro handicapper or someone who is a very serious amateur. Regardless, if you've been making money for years, you've been playing with an edge. Therefore, I'm assuming that you have some method of determining that you have an advantage in a given handicapping situation from your experience with similar past situations. So you are already reflecting backward on your gambling activity to make bets. As you know, this is exactly how Wall Street works - they project future bets (investments) based on past performances (historical stock data). From what you write, I'm also assuming that you're not a database handicapper, and thus don't have the ability to data mine your records (at least conveniently) for profitable factors and angles, but rely more on your memory. Correct me if I'm wrong. It's very possible that the factors you use to evaluate bets are more stable than you think. If you do in depth record-keeping, you might be able to profitably datamine you records.

The value of Kelly is that it enables the bettor to use the degree of precision of his (or her) edge (assuming they have one) to create optimal bankroll growth. Not only do all professional blackjack players use Kelly, but it's widely used in the financial world, particularly by hedge funds. Warren Buffett is probably the most famous Kelly advocate, and he hasn't done too badly.

Cheers,

lansdale

[lansdale]

Originally Posted by Lefty

Huey did more work with Kelly than anyone I ever met or read about.
It's no more dangerous than flat betting. You set a bankroll and don't go
crazy.

Horse racing percentages fluctuate, that's why Huey updated it regularly, so as to make the necessary adjustments. I think that's the way not to get hurt instead of having a model you're hoping will stay stable.

My model didn't stay stable and that's why I bailed out. Precisely because I did update the model every 15 races is why I could still bail with most of the money.
Edges change constantly and very likely there will come a time where they will
disappear altogether. I think to think you will have a constant non-changing edge is insanity.

Hi Lefty,

An edge can only be defined by some measure of stability. In blackjack there is a measurement known as N0 (not NO), which is the number of hands required to double one's bank. 50k is the number of hands required for a benchmark game. Since blackjack is a mathematical system, your edge, assuming given conditions, won't change in a million years.

But blackjack has much higher variance than horseracing, and provides an edge of only approximately 1%. Although horseracing may seem more volatile to many handicappers, the game can offer lower variance and a much higher edge, to knowledgeable players, especially with rebates. Thus, smaller samples of races can be used to create viable models. Many successful handicappers, including the celebrated William Benter, have suggested that somewhere between 2k and 3k races are a sufficient sample size to test racing data. 'sjk', an occasional poster on this site, has created a model than has been stable for over a decade with barely any tweaking. What you're describing as the 'changes' of the game is simply normal variance. If your model couldn't handle this, it wasn't a viable model.

Cheers,

lansdale

[Dave Schwartz]

But blackjack has much higher variance than horseracing, and provides an edge of only approximately 1%.

Actually, horse racing has the higher variance.


But I agree with most of what you have said.