BB Food for Thought

[SoCalMan2]

miff Wrote:
-------------------------------------------------------
> \"Third, any player is going to lose more
> regression plays against heavy favorites than they
> win. There is no question about that. However, if
> you can win 50% of these plays, then, with the
> likely fat returns, you are really doing
> superbly\"
>
>
> Hi So Cal,
>
> The most successful players in the country do not
> win anywhere near 50% of their plays unless they
> are playing filtered short priced horses.Did you
> mean 5% of the time?
>
> Historically,odds on 2/5 shots win at more than a
> 50% rate, so probability wise it\'s not possible to
> cash 50% betting against all of them.Maybe you
> meant you are looking at strictly filtered
> regression plays,like the one you see here against
> BB.After BB and CD,I think the odds will be all
> overlays, so the possibility of a monster payoff
> does exist if both run out or if BB runs out.
>
>
> Mike


Hi Mike!

You are correct. 50% was a bad number for me to use. I was just trying to explain (albeit poorly) that there are plenty of cases where a bettor can make very good bets where chances are high he/she will not cash.  The classic example is if you are able to find 100-1 to win on a horse that has a 10% chance of winning (assume we know the 10% chance is correct).  Obviously, the bettor will lose that bet 90% of the time.  However, the bettor is unconcerned about losing 90% of the time (and in fact expects it to happen) because he/she only needs to cash every once in a while (one out of ten is just fine) to show a very healthy profit.  Unfortunately, nothing is ever as simple as that example, and I expect that constructing the correct play will prove very challenging.  Also, one needs to make sure his/her bankroll can withstand the inevitable streaks of a lot longer than 10 times without cashing because randomness is naturally streaky (as strange as that sounds).  In any event, you caught me on a bad piece of reasoning.  Thanks for keeping me honest.

SCM2

[miff]

So Cal,

Understand your methodology and thinking. Very true that strict exotic/bomb players must be able to withstand long droughts until they hit. Good Luck

Mike

[TGJB]

Coupla things. First of all, no, I do not think it would be good for the game if we have the first TC winner in 30 years, and it\'s trained by Dutrow.

Second of all, I agree that nobody can tell with 80% certainty what BB (or any horse) will do-- that\'s why he\'s a bet against at 3-10 or whatever, as So Cal says. As you said (in effect) when going through your Derby analysis, it\'s about percentages, not picking winners.

Third of all, not specific to you, but to this discussion in general:

THE LAST ELEVEN OF THESE HAVE GONE DOWN. Not one, not  two, eleven in a row. There are reasons for that (fkach, leave it alone). Sure, this could be the one, and maybe he has a better chance than most of putting in another one just because of Winstrol, etc. But it can\'t be wrong to take a shot against him, given the odds vs. the history alone-- even if you don\'t look at our data.

[fkach]

>But it can\'t be wrong to take a shot against him, given the odds vs. the history alone-- even if you don\'t look at our data.<

I don\'t see anyone advocating betting on him. I think it\'s universally accepted that he will be an underlay. It\'s how much of an underlay that\'s debateable.  Just because his expectation is for a negative ROI, that doesn\'t translate into profits elsewhere unless he\'s a \"huge\" underlay.

In order to tell whether it\'s correct or not to take a shot against him, you have to be able to evaluate the chances of his form deteriorating, not being able to get 12F, getting an unforeseen disaster trip, or someone moving past him. Then you have to compare the probabilities to the odds.

I think the real discussion is trying to estimate the probabilities of the various negative scenarios. IMO the only way to do that is to look at the individual horse in question and not the failures of other horses.

[BitPlayer]

SCM2 -

Always enjoy your posts.  I was struck by the similarities between your inital Belmont post and all the discussion of \"bubble\" economics that one reads these days.

I have a comment and a question.

Comment:  You posted \"There were plenty of people who thought he was going to regress in the Derby. Whoever thought that has to think the chances of a regression are greater now.\"  I don\'t think that\'s necessarily true.  When a horse puts up one big number, there are three explanations I can think of:

(1) He\'s really good.

(2) He had a really good day.

(3) The figure is off.  (I know it\'s rare, but it must happen occasionally.)

Explanations (2) and (3) will both generally lead to a regression next time out and had to be considered for Big Brown going into the Derby.  After Big Brown put up a second big figure in the Derby, those explanations became much less plausible, and those looking for a regression must now rely on other reasons for one to occur.

Question:  Based on the information you now have, how likely (in percentage terms) do you think it is that Big Brown will win the Belmont?

[TGJB]

Bit-- your reasons leave out the entire reason we put the figures on a graph (CTC thinks this way too). We can think that a horse will regress BECAUSE he ran \"really good,\" or because of the cumulative effects of really good efforts.

[BitPlayer]

TGJB -

I understand.  In my post, that would fall into the category of other reasons for regression to occur.

My point was simply that one-hit wonders (like Barbaro, Bandini, and Bellamy Road), which is what Big Brown was prior to the Derby, aren\'t in the same category as horses who have shown the ability to replicate a top effort, which is what Big Brown was after the Derby.  In my mind, the salient examples going into the Belmont are Afleet Alex and Smarty Jones.

[TGJB]

AA wasn\'t coming off big efforts in the Derby and Preakness. SJ is the closest thing we have to a model, though there are differences.

[Beginner]

I was initially betting very small amounts on races and recently started deploying a bit more capital.  

What is glaringly apparent, however, is that with a bit more knowledge on how to efficiently construct a bet once analysis is complete, I could be doing significantly better.  I\'ve read and learned an ENORMOUS amount of information from this board and am ever grateful for the education.  When I first started reading the Board I was on google every other minute looking up stuff - didn\'t know Fager from Furlong.  I still have a LONG way to go.  

One thing that I haven\'t seen yet (and is most likely here somewhere) is a discussion on constructing a wager.  I read with great interest MJ\'s post on his Derby analysis:  What I wasn\'t sure is how you figured the $7K layout - was it a function of a predetermined bankroll, or was it simply the total of your bets?  In races like the Derby, Preakness, Belmont, BC, etc., the pools are orders of magnitude greater than an ordinary day of racing.  It follows that those are the races where it would seem to make sense to ratchet up one\'s typical bets.  

What I haven\'t quite wrapped my brain around is the efficiency of an exotic strategy vis a vis the size of the pools and separately, some of the vanilla payouts.  Does anyone have a good resource or advice on this topic?  I think fkach started to discuss something a couple of days ago related to place / show pools and overlays that I didn\'t quite understand - any further comment would be greatly appreciated.  Thank you!

[fkach]

I was pointing out that on big race days there are much larger betting pools partly because a lot of unsophisticated money gets bet. That creates more inefficiencies (overlays) than you typically see on a day to day basis. The reason I like to look at the place and show pools on days like that is because the sophisticated big money bettors do not look for overlays in those pools. So the odds do not get corrected by late betting. THey stay at overlaid prices.

Basically you need to learn how to translate a horse\'s chances of winning into his chances of placing and/or showing. Then you have to compare those chances to the payoff range on the horse to see if you are getting good value. If the lowest possible payoff is about break even, then I start getting interested because very often I\'ll get a bigger price than that if my horse comes in with a longshot.

I don\'t know how much you know about calculating place prices etc... but it\'s a little tricky these days because of \"Net Pool Pricing\".

Other than that, ask a specific question or contact me via PM.

[hossgnat]

A couple of years ago Steve Crist filled a gaping void in the handicapping/horseplaying literature with \"Exotic Betting\".

It\'s a wonderful treatise on constructing exotic wagers to represent your genuine opinion of a race or races, regardless of the size of your bankroll or level of expertise.  Plenty of other interesting insights too.

[mjellish]

Wager Construction / fkach/MJ (75 Views)
Posted by: Beginner (IP Logged)
Date: May 22, 2008 04:09PM


I was initially betting very small amounts on races and recently started deploying a bit more capital.

What is glaringly apparent, however, is that with a bit more knowledge on how to efficiently construct a bet once analysis is complete, I could be doing significantly better. I\'ve read and learned an ENORMOUS amount of information from this board and am ever grateful for the education. When I first started reading the Board I was on google every other minute looking up stuff - didn\'t know Fager from Furlong. I still have a LONG way to go.

One thing that I haven\'t seen yet (and is most likely here somewhere) is a discussion on constructing a wager. I read with great interest MJ\'s post on his Derby analysis: What I wasn\'t sure is how you figured the $7K layout - was it a function of a predetermined bankroll, or was it simply the total of your bets? In races like the Derby, Preakness, Belmont, BC, etc., the pools are orders of magnitude greater than an ordinary day of racing. It follows that those are the races where it would seem to make sense to ratchet up one\'s typical bets.

What I haven\'t quite wrapped my brain around is the efficiency of an exotic strategy vis a vis the size of the pools and separately, some of the vanilla payouts. Does anyone have a good resource or advice on this topic? I think fkach started to discuss something a couple of days ago related to place / show pools and overlays that I didn\'t quite understand - any further comment would be greatly appreciated. Thank you!



Hey Beginner,

Not sure how much this will help, but here goes.

I guess the short answer is this. This was not a pick 6 type bet, in which case you sometimes have to play at least $xxxx amount to feel you have a reasonable chance of winning.  So my bet size was determined first by my confidence in my position & second by the amount of capital I was comfortable losing.  I will usually play about $200- $300 on any race that I have some type of position.  I may go as much as $1500 if I like my position and am getting very good value.  5k is usually about the absolute most I will bet on a single race if I have maximum confidence and am getting ridiculous value.  In this case it was the derby so I reached a bit and stretched to 7k.  I will bet more into a pick 6, pick 4, multi race gimmick, etc, but only if I feel I have a strong position (meaning a single that is not a favorite, or better yet a long shot, and also preferably another race or two in the sequence where I like a few bombers and/or hate the favorite - and of course there has to be a carryover or guaranteed pool).

So in the case of the derby, I decided what I was comfortable betting and then structured what I thought was the optimal way to bet the race.

One word of advice.  IMO most people play this game exactly the wrong way.  For example, they like a horse and make him even money to win.  He goes off at 8/5 so they bet $200.  If he goes off at 3-1 they bet $100, and if he goes off at 5-1 they still bet $100 or maybe even less.  Right way to play this is the opposite.  If you like a horse and he goes off at 5-1 you should bet more on him than you would if he went off at 8/5.  That way you are putting your money on value, which is the key to this game.  If you like a horse and he offers no value (like taking 5/2 on BB to Win the derby), see if you can find a way to get value by going to the exotics and muti race exotics.
 

Hope this helps.

Happy Hunting,

MJ

[Chuckles_the_Clown2]

He just was not remotely in the same speed ballpark as the faster horses. What was remarkable was that he went off at nearly the same price Street Sense did. You gotta love the folks that bet down horses upon conjecture. It took some doing, but I finally talked Steve Haskin off him.

I have some statistical research for our friend Rags to Richie. Which trainer has the most last place finishers in the Derby?

He would have to really suck wouldn\'t he? I mean stink to high heaven and be without a shred of horsemanship and never know when to run a horse and when not to run! He would be a charlatan.

2008-Monba
2007-Cowtown Cat
2006-Keyed Entry
2005-Bandini (2nd to last)
2004-Pollard\'s Vision (2nd to last)
2002-Wild Horses
2000-Graeme Hall

This guy has 5 last place finishes and 2 second to lasts. I don\'t think anyone has ever done worse than that.

Silver Charm Wrote:
-------------------------------------------------------
> Very interesting the path and free pass Pyro is
> taking and has gotten. For a horse who was so
> widely regarded as \"THE HORSE TO BEAT\" in the
> Derby as late as late March he is now totally off
> the Triple Crown Trail and not even contending in
> the second tier 3YO races.
>
> He burned a lot of money Derby Day and now he is a
> Northern Dancer Stakes, Barbaro Stakes type horse.
>
>
> Hmmmmmm

[imallin]

Very good post, let me chime in with my thoughts.

BB\'s Beyer fig in the Derby was 109 and his Preakness number was 100. The way people are talking, Big Brown is being hailed as an all time great. Commentators like Gary \"conflict of interest\" Stevens and laffit Pincay III have talked in terms that i don\'t think i\'ve ever heard used for a thoroughbred. I can\'t imagine that any commentator in 1973 ever used MORE glowing terms for Secretariat.

I haven\'t heard one person anywhere in the world mention that Big Brown is anything less than the greatest thoroughbred of all time. \'Secretariat who\' is what everyone is saying.

So, if Smarty Jones ran 118 in the Preakness and BB ran a 100, what am i missing?

Isn\'t an all time great thoroughbred supposed to run 120 or 125 in at least ONE of their lifetime starts? Didn\'t Mineshaft and Ghostzapper run 120 or faster on the Beyer scale?

[SoCalMan2]

BitPlayer Wrote:
-------------------------------------------------------
> SCM2 -
>
> Always enjoy your posts.  I was struck by the
> similarities between your inital Belmont post and
> all the discussion of \"bubble\" economics that one
> reads these days.
>
> I have a comment and a question.
>
> Comment:  You posted \"There were plenty of people
> who thought he was going to regress in the Derby.
> Whoever thought that has to think the chances of a
> regression are greater now.\"  I don\'t think that\'s
> necessarily true.  When a horse puts up one big
> number, there are three explanations I can think
> of:
>
> (1) He\'s really good.
>
> (2) He had a really good day.
>
> (3) The figure is off.  (I know it\'s rare, but it
> must happen occasionally.)
>
> Explanations (2) and (3) will both generally lead
> to a regression next time out and had to be
> considered for Big Brown going into the Derby.
> After Big Brown put up a second big figure in the
> Derby, those explanations became much less
> plausible, and those looking for a regression must
> now rely on other reasons for one to occur.
>
> Question:  Based on the information you now have,
> how likely (in percentage terms) do you think it
> is that Big Brown will win the Belmont?


Bitplayer,

Thank you for the compliment.  As to your question, I would say 20-25% is how likely I think it is that Big Brown will win the Belmont.  When I can find the time to explain, I will.  However, that is a very rough estimation at this point.  There are a ton of variables that can change things.

SCM2