Man Bites Dog

[BitPlayer]

If you took 4-to-5 on a coin flip, bet heads, and it came up heads, would you say you had made a good bet?

[TGJB]

No horse is ever 100% to win, so it\'s a game of percentages.

To choose a simple example-- two horses in a ten horse race. One is the single most likely winner, he\'s 30% to win. The other is 15% to win. The first horse is even money, the second 10-1. You are more likely to \"hit\" the race if you bet the first horse. But man, will you lose money in the long run making that bet. And in the long run you will make money betting the other one, even though you are half as likely to cash any given time.

[TreadHead]

Have no idea if this tweet is correct, but quite possible that Dan does have access to Rags/Jake.  What\'s confusing is he calls it a \"Giacomo like 7\" but Giacomo ran a 5 on Rags.  If it is a 7, that would make it the slowest one in 25 years tied with California Chrome (wait...what?)

https://twitter.com/fatbaldguyracin/status/993999193151889408

[TGJB]

It’s interesting , they haven’t posted the figures on their site as they usually do...

[Wild Again]

Unless I am very mistaken Instilled Regard was given by someone on this board as a wet horse move up.

So the discussion of Instilled Regard may be pertinent.

Just sayin. And if it was never mentioned and I imagined it, I apologize in advance.

Regarding the 4-5 coin toss bet, read up on the idea of Gambler\'s Ruin.
Your statement is not as simple as you make it out to be.

The Kelly criteria was theorized invented choose whichever word is best for you, to explain a set of circumstances where betting plus expected value can still lead to the Gambler\'s Ruin.

The problem with horse racing is it is almost impossible to have enough information to know 100% the correct amount to bet.

Or in other words it is very hard to maximize your play.

To the thorograph board this may be an assertion as I have presented neither the science nor the history of the Kelly criteria but if you are really that curious I suggest you do your own research.

You might be pleasantly surprised.

Thanks

John Perona

[ajkreider]

The issue is betting value vs. expected value given the evidence. That goes for things like looking at sires to figure out who may move up in the wet.

Yeah, you could \"make a case\" for an IR move up, and it was in fact made.  You could make a case for a bunch of horses running a better number than IR as well.  That IR ran well in the Derby does not make using the sire data as reason to bet him over others.  It\'s a data point to put into play with others, including his odds.  

JB is exactly right, and it has nothing to do with \"gambler\'s ruin\".  If you\'re looking for a sure fire thing, get out now.  Even Dutch book cases can lead to ruin for the casino, if you get a bad run. You won\'t have perfect evidence ever.  It\'s just about putting yourself in the best position to win.  I will book the correct side of the Monty Hall paradox all day.  I  could lose, but I wouldn\'t bet on it.

Play some hold\'em.  You will get beat, all in, on the river, because the other guy lands his only out.  If losing the right bet is something you can\'t stomach, then you shouldn\'t bet at all.

[P-Dub]

johnnym Wrote:
-------------------------------------------------------
> Statement was make the right bet not pick the
> winner.
> If you are strictly a win better it’s irrelevant
> what the odds are if you can’t pick the winner.
> Just saying

I understand the statement.

Your second sentence is irrelevant to the discussion.

[jbelfior]

What if Drayden goes around the wall of 5 that was stopping in front of him instead of deciding to wait and weave through them?

Smarter people,with better eyes than me, put IR at least second if DVD makes a better decision at the quarter pole.


Good Luck,
Joe B

[johnnym]

I understand that is not a smart wager as there are only 2 outcomes and I should at least get those odds.

[Paolo]

Is that really what I said? Not a great reading on your part, TPeck. I asked for thoughts AFTER the race. Absolutely nothing to do with wagering. No sour grapes either, since I got the sheets but not the seminar. Seems like all but a few on this board think of their responses before they even read the GD post.

JB graciously provided his POST-Derby numbers and as I looked them over, I saw two horses that deserved discussion. Not sure how solid the figs could be considering the sloppy track and hot pace. I was obviously wrong. The two jumps to zero deserve zero discussion from the host, literally. Guessing the other guys didn\'t find it so easy. But rest easy, JB is on the case of the Rags delay.

[Wild Again]

I would like to ask a hypothetical question.

Assuming you have a perfect odds line 100% correct.

You are faced with a 20-1 who you believe wins the race 1 out of 10 times.
Or in other words, the horse is 20-1 but should be 9-1.
Also assume the odds will not go lower after you have made your bet and all the other horses in the race are fairly priced.

How much do you bet?  As a % of bankroll and as a multiple of your bet size?

As I believe this is the most important wagering question. Any help would be appreciated. And is a question I haven\'t been able to answer in 45 years.

Thanks

John Perona

[Furious Pete]

Do a little search for Kelly Criterion in the forum archives, and you can see that MathCapper has answered this in great extent multiple times. If you believe this is the most important question in wagering, you should really check that out. You can also use 1/2 Kelly, 1/4 Kelly as per your own preferences - but MathCapper has this covered. Full Kelly is very aggressive, and might very well lead to \"gamblers ruin\" as I now see you referred to in another post. In simplistic terms, with Kelly Criterion you increase the percentage of your bankroll wagered as your edge increases, and as the odds decreases, thus answering your very hypothetical question.

I believe one could get in the ballpark by just internalizing this as a simple rule of thumb, let me rephrase it as a rap lyric: \"You can tap it hard\' when you dig your edge, n\' if the odds is cheap, don\'t hedge\".  

A simple way to combat gamblers ruin is to play with a fractional Kelly, and then withdraw a good chunk of your bankroll after you reach your first zillion or so, maybe buy a cruise ship or something for the money and continue gambling with the rest. Then you wouldn\'t need to worry about losing it all in a bad streak, and you could even trade your cruise ship in to a lesser boat if you still want to continue betting after that horrible experience of losing your remaining bankroll. (the example is scalable)

[BitPlayer]

Mathcapper had a helpful post about this a few years ago.

https://www.thorograph.com/phorum/read.php?1,82455,82455#msg-82455

[prist]

Your bet size depends on your edge and bankroll.

in the example you posted your edge is [1.1]
if your bank is $100 the wager is $5
if your bank is $1000 then the wager is $55

It\'s a difficult way to play because you don\'t know the final odds in advance. I used to use the conditional wagering option at TwinSpires, but I received an e-mail from them just this past week indicating they will no longer offer this wagering service.

[TempletonPeck]

After the race, we\'re all weak to hindsight bias, and to learn something meaningful, it\'s important to have/look at a big data set so that you aren\'t cherry-picking to support your ideas.

For example, if you look at the sheets of either horse you are discussing here, you may be able to say \"Ah, sire X/trainer Y clearly has horses move forward first time slop in the Derby!\" But, if you looked at 10,000 of each, I think you\'d more likely find that isn\'t true. (Of course, we don\'t have 10,000 of anything to look at here, but we may as well look at as many as we have!)

Looking at individual performances may be a good enough place to come up with an idea, which you then use data to confirm or deny or support or what-have-you, but that\'s all it\'s really worth.

So, as I have previously suggested, if you want to look through the (free!) archives and say \"Folks, I have done some digging, and I have an idea about horses that tend to move up in sloppy Derbies:\" I think you would get an absolutely massive and constructive response. But, when you say \"I think your time would be better spent figuring out why you didn\'t know these horses would move up, rather than worrying about your competitors,\" I think you should expect a response in kind.