Best Betting Strategies for Long-term Success?

[castaway01]

Originally Posted by thaskalos

The "proper" betting strategy depends on the individual player's temperament and personality. It seems strange to me that most of the horseplayers -- from what I see on this board at least -- appear to favor win-betting...because win-betting goes against the temperament of most horseplayers with whom I have become acquainted. Almost all the horseplayers that I know hate losing streaks, and once they fall behind in their betting they get very impatient and look to get even right away...and win-betting goes against this sort of mentality. The win-bettor must be patient and resolute...and these aren't personality traits that are commonly found in the typical horseplayer.

More important than what betting strategy we should use is the question of what sort of person we are as far as temperament and personality are concerned. Because if our betting style doesn't match our "inner horseplayer"... then our chances of surviving in this game hover around zero.

For your first paragraph, I think it's just the difference between what is posted here, which is how we would want to play in a hypothetical best-case scenario (efficient win betting nailing price horses all day) and what we see in real life, which is bettors losing and then chasing with exotics.

[Running Amok]

Originally Posted by MJC922

Correct. I'll upload some screenshots for you. First one is the race before the Haskell, matrix highlighted from Amwager. Second one is showing it pasted into my Excel sheet, projected odds. Third image is the actual result with the final odds highlighted.

For a better understanding of the process you can google "doc constructed a situation" and that will bring up google books hopefully, then see page 138 of the Winning Horseplayer, (google for now if you don't own it... but we all should on this board anyway)

Thanks for the screen shots. Your projected odds were pretty damn close, impressive.

I have two of Andrew Beyer's books and one of them, Beyer On Speed, he autographed for me during a promotional book signing day at Detroit Race Course back in the early 90s.

So on page 138 in the Google book version of the Winning Horseplayer, he's talking about exacta prices. I get the idea behind what he's doing and it makes sense. But that doesn't really relate any info on the process of projecting odds from a DD matrix.

Did you design that Excel spreadsheet and write the formulas yourself?

[Dave S]

Did some looking and found this. Scroll down to find the link to the spreadsheet. This uses the will-pay data, but apparently you can use other multi-race data besides just the DD. https://inthemoneypodcast.com/imputedodds/

[Running Amok]

Originally Posted by Dave S

Did some looking and found this. Scroll down to find the link to the spreadsheet. This uses the will-pay data, but apparently you can use other multi-race data besides just the DD. https://inthemoneypodcast.com/imputedodds/

Thanks Dave S.

Yes I had already found that and tested it. While it doesn't do a bad job of projecting the odds, it's not the method what MJC922 is using. He's using the entire matrix for a more accurate projection. Because as he stated earlier:

Willpays are the race winner's probables from the first leg, nothing more. If that's all someone is using it means they're looking at just a slice of the wagering that was actually done.

You want to process all of the combinations to get the best projected odds for yourself.

[MJC922]

Originally Posted by Running Amok

Thanks for the screen shots. Your projected odds were pretty damn close, impressive.

I have two of Andrew Beyer's books and one of them, Beyer On Speed, he autographed for me during a promotional book signing day at Detroit Race Course back in the early 90s.

So on page 138 in the Google book version of the Winning Horseplayer, he's talking about exacta prices. I get the idea behind what he's doing and it makes sense. But that doesn't really relate any info on the process of projecting odds from a DD matrix.

Did you design that Excel spreadsheet and write the formulas yourself?

Thanks but I can't take any credit for getting close, the crowd gets that credit. It makes perfect sense that it will be close because it's a finalized pool for the race. At NYRA it's generally spot-on. At least the tracks that I tend to play it's plenty close enough for my purposes.

As far as the process I can't go any further than what's been posted. If you copy and paste the matrix into an excel sheet (or any spreadsheet for that matter) and play around with it for a while then I'm sure you'll eventually get to the percentages you need here, then simply apply the takeout and convert into decimal odds. It may not look all pretty but that's not necessary anyway. Mine has a custom function or two that I'd rather not post but if folks want some guidance on making pretty or more functional, I can assist some more if needed. I know there can be some time to wrap ones head around it but it's just a matter of pasting it into excel and messing around with it until it eventually clicks. Even if that takes someone a couple of weeks or a month I'm confident it'll be the best time you've ever spent. Keep in mind you're just dealing with payoffs, the fact that they're exacta in the book vs double in this case doesn't matter. At the end of the day it's a bunch of percentages is all.

[Onesome]

Originally Posted by Dave S

Did some looking and found this. Scroll down to find the link to the spreadsheet. This uses the will-pay data, but apparently you can use other multi-race data besides just the DD. https://inthemoneypodcast.com/imputedodds/

tyty been looking for something like this forever.

[Dave S]

No problem. From the limited look that I have seen it seems to be in the ballpark with its data.

[Running Amok]

Yes, its looks close to the same as using the full matrix. And since this is just an estimate, perhaps just using Will-Pays is all that's needed. But why not try and ge the most accurate estimate possible?

Although I never asked, he has said he won't post the formula. I don't blame him and that's his prerogative. Besides, that would be too easy!

I can already see how he gets the numbers in the bottom of the matrix.

In the top matrix in column C the DD pays 351 for the 1/1 combination.
That is for a $2 DD. The $1 DD paid $175.40

So in the bottom matrix the result is 2.85
That result is simply the 1/the payoff 1/351 = 2.85
Actually the result is .002849 but he moved the decimal over to the right 3 places.

You could also divide 1000/the payoff 1000/351 = 2.85

So c'mon guys, lets try and crack the code of the matrix posted by MJC922.

HaskellProjectedOdds02.jpg

[Running Amok]

To make things easier I recreated his spreadsheet.

I already entered the formulas to populate the bottom matrix

Lets try and figure out the formula that projects the odds

I formatted the cells for odds as text and typed in the correct results in red.

I have a feeling MJC922 isn't showing us the entire spreadsheet.

He probably has some formulas and calculations in other cells.

I also think he probably got the idea and some of the formulas from the link shared above for the spreadsheet that uses Will-Pays to project final odds.

MJC922 DD Matrix.xlsx

[MJC922]

Originally Posted by Running Amok

To make things easier I recreated his spreadsheet.

I already entered the formulas to populate the bottom matrix

Lets try and figure out the formula that projects the odds

I formatted the cells for odds as text and typed in the correct results in red.

I have a feeling MJC922 isn't showing us the entire spreadsheet.

He probably has some formulas and calculations in other cells.

I also think he probably got the idea and some of the formulas from the link shared above for the spreadsheet that uses Will-Pays to project final odds.

Attachment 33214

For the record I've never once seen the other spreadsheet at that link and it would not have interested me much in the first place due to the limitations of will pays. Anyway I've been hearing will pays on this board for about 15 years and I'm very familiar with the level of accuracy they provide, it never was particularly good but I do agree that it's far better than nothing so if it's good enough for most folks I'm happy someone has shared that work. Until I started using the full matrix I was thinking fixed odds wagering would be the only way I'd be taking the game seriously again but I'm confident fixed odds aren't necessary (which is a good thing because I can't relocate to NJ). Nice to have fixed odds as an option though. As for my sheet and the formulas I mean there really aren't any, it just follows what Beyer described pretty much 1-1, i.e. 1000 divided by payoff, add it up and divide. I'm not trying be sly with this. It's just about reading Beyer and adapting it for doubles. If you can add and divide that's about it and a custom function to go from decimal to fractional odds to 1. That's only cosmetic.

[Running Amok]

Originally Posted by MJC922

As for my sheet and the formulas I mean there really aren't any, it just follows what Beyer described pretty much 1-1, i.e. 1000 divided by payoff, add it up and divide. I'm not trying be sly with this. It's just about reading Beyer and adapting it for doubles. If you can add and divide that's about it and a custom function to go from decimal to fractional odds to 1. That's only cosmetic.

I never thought you were being sly. I figured you were guarding your work.

As you can see, I've already figured out how to get the numbers in the bottom matrix which = 1000/payoff.

I am not interested in the cosmetics. I already use a custom function to convert fractional odds to 1 in my spreadsheet, too.

But to get to the fractional odds, after adding to get the total of the 1000/payoffs values, what value is used to divide by?

[Running Amok]

OK, I got it! I figured it out. Like you said "It's just about reading Beyer and adapting it for doubles."

It took me a while to make the connection but certainly glad it didn't take weeks or a month

After I played around with the numbers some, like you said it just clicked.

That Exacta example in Beyer's book definitely held the key.

Thanks for all the help and posting the phrase to help find his book on Google.

The beauty of this is no longer being concerned with the late tote odds shifts.

This is a game-changer and processing all of the combinations instead of just the will-pays gives me confidence I'll be armed with the best projected odds I can make.

I have been frustrated in the past with changing odds and I'm hoping using projected odds will make the CAW whales irrelevant.

[MJC922]

Originally Posted by Running Amok

OK, I got it! I figured it out. Like you said "It's just about reading Beyer and adapting it for doubles."

It took me a while to make the connection but certainly glad it didn't take weeks or a month

After I played around with the numbers some, like you said it just clicked.

That Exacta example in Beyer's book definitely held the key.

Thanks for all the help and posting the phrase to help find his book on Google.

The beauty of this is no longer being concerned with the late tote odds shifts.

This is a game-changer and processing all of the combinations instead of just the will-pays gives me confidence I'll be armed with the best projected odds I can make.

I have been frustrated in the past with changing odds and I'm hoping using projected odds will make the CAW whales irrelevant.

You're welcome and good job.

[Running Amok]

I just wanted to mention to anyone else trying to use this method for projecting the final odds. I noticed a few typos in the spreadsheet I posted. See below

In the spreadsheet I posted, for horse #6 the payouts with horses 1,2 and 3 in the top matrix are wrong.

I accidentally copied the first three payouts listed for horse # 5 under #6. And this will cause the calculations in the bottom matrix where 1000 is divided by the payouts to be incorrect.

So referring to the screenshots posted by MJC922 the correct payouts for horse #6 with the 1, 2 and 3 are 2137, 5364 and 477, respectively.

Originally Posted by MJC922

You're welcome and good job.

Thanks again. Would it be OK to send a PM?

[Poindexter]

MJC, question on how you are handling 2 scenarios.

Scenario 1) You make a horse 3-1. You demand 9/2. Horse shows 4.6-1 as his daily double projected odds. So you have have a potential play. Horse is 7/2 as they are loading. Do you bet him thinking he will become 9/2 after off? 2nd scenario, you make a horse 3-1, he is only 3.6-1 on daily double projected odds. No play. But surprisingly he is 9/2 when they are loading the gate. Do you bet him or do you pass, assuming he will be hit late in the betting. Also you mentioned NYRA, where caw is heavily is involved in the double pools not nearly as involved in the win pool. Logically there would be a significant difference in the daily double projected win odds and the actual win payouts unless a lot of folks are using the double pools to find value in the win pools.