Equine vs equity investing

[gm10]

" Is betting on horses very different from picking stocks? Can understanding a gambler's approach and mentality give a better understanding of fund managers?

In searching for answers to these questions, I spoke to Paul Moulton, a professional gambler who originally worked in the fund management industry."

http://www.reuters.com/article/2011/12/23/lipper-column-idUSL6E7NM1MG20111223

[FenceBored]

" Is betting on horses very different from picking stocks? Can understanding a gambler's approach and mentality give a better understanding of fund managers?

In searching for answers to these questions, I spoke to Paul Moulton, a professional gambler who originally worked in the fund management industry."

http://www.reuters.com/article/2011/12/23/lipper-column-idUSL6E7NM1MG20111223

Money graph:

http://graphics.thomsonreuters.com/11/12/GLB_LPPRCLM1211_VF.jpg

[erikeepper]

Nice article! Thanks :)

[Overlay]

"There's no point in picking winners if you're backing them at the wrong prices, because sooner or later you'll come a cropper."

So, all you non-value players, remember: Don't come a cropper! (I love British English!)

[CincyHorseplayer]

Thanks for the article GM.Good read.The guy sounds a lot like Raybo who has the exact win % in superfectas and is a profitable.

[classhandicapper]

Money graph:

http://graphics.thomsonreuters.com/11/12/GLB_LPPRCLM1211_VF.jpg


I think this is a great topic for discussion.

My own view is that the more information you have the better off you are.

The problem is the more information you have, the more difficult it gets to weigh it all and assess each horse's chances of winning properly and determine if you have good value.

For example:

If the only thing you knew about a horse was that it raced on a dead rail 3 times in a row and the 3 races before that were good enough to win today, you'd probably already know enough to have good value in most cases.

But suppose two other horses had a positive trainer pattern going for them and another horse had pace figures that suggested it could shake loose.

If you knew all that, you might be totally confused about what the appropriate prices were.

[windoor]

I looked at a play earlier this year that only had a 10% hit rate. The play required that you wait for an odd of ten to one or higher. I have had some serious losing streaks with a 27% hit rate. I can just imagine what it would be for one that only hit at10%.

Still, it looked to be viable long term if you have the stomach and bankroll for it.

"For Moulton it is in the nature of things that he will hit bad spells as winners don't come neatly spaced between losses. His approach requires consistency but a run of losers has often tempted him to start tinkering with that approach in order to try and end the losing streak - a fatal error."

I have done this many times, until I learned my lesson. Another hard one. The bank needs to be much stronger than you might think, and we are back to patience and discipline. Funny how that keeps coming up in the final analysis.

Regards,

Windoor

[cj]

I think this is a great topic for discussion.

My own view is that the more information you have the better off you are.

The problem is the more information you have, the more difficult it gets to weigh it all and assess each horse's chances of winning properly and determine if you have good value.

For example:

If the only thing you knew about a horse was that it raced on a dead rail 3 times in a row and the 3 races before that were good enough to win today, you'd probably already know enough to have good value in most cases.

But suppose two other horses had a positive trainer pattern going for them and another horse had pace figures that suggested it could shake loose.

If you knew all that, you might be totally confused about what the appropriate prices were.

I like the less info model. If you have information that you know is always underbet enough to produce long term profits, that really is enough.

[Robert Goren]

The article is about gambler in the UK. There is world of difference in the way horses are bet there and here. There you have price certainty when you make a bet. Things get a lot easier when you know exactly what the return is when you make your wager. That goes double when your profit margin is in the single digits.

[badcompany]

"So in 2010 when he sold his beloved E-Type Jaguar to raise some cash, he did not adopt a more cautious approach and proceeded to lose the money made from the sale in 45 minutes on the racecourse."

Sounds like a guy who will go bust before long. He had a hot year and went out and bought an expensive car assuming that year would be the norm.

One of the cardinal sins of money management is to show a disregard for money when things are going well.

[Fingal]

I like the less info model. If you have information that you know is always underbet enough to produce long term profits, that really is enough.


When something simple works & you strive to make it better, one usually makes it worse.

[LottaKash]

I like the less info model. If you have information that you know is always underbet enough to produce long term profits, that really is enough.

I ditto that....It took me a very, very long time, but that is pretty much where I am in all of this, these days..."Less is more"

It limits you a bit, action wise, but I was quite surprised in more recent times, not as much as I had suspected without some tangible proof, which is the +/- $$$ at year's end...

best,

[classhandicapper]

I like the less info model. If you have information that you know is always underbet enough to produce long term profits, that really is enough.

I tend to do better this way also, but I also think it's possible to weed out bad plays within the group with more information.

Take my bias example above.

A horse whose form is that hidden will probably be profitable long term, but if someone else is going to control the pace and two other horses might move up a lot because of a trainer move, perhaps you are better off just passing that race.

That's more or less where I'm at.

I sort of stopped trying to weigh it all as much as I used to. I just want 1 or 2 things going my way and nothing going against me.

I actually picked that up style betting basketball. When I first started betting basketball, I knew I couldn't weigh all the contradictory factors in my spreadsheet well. So I only bet when I had 2 things going for me and nothing going against me and it worked really well.

[cj]

I tend to do better this way also, but I also think it's possible to weed out bad plays within the group with more information.

Take my bias example above.

A horse whose form is that hidden will probably be profitable long term, but if someone else is going to control the pace and two other horses might move up a lot because of a trainer move, perhaps you are better off just passing that race.

That's more or less where I'm at.

I sort of stopped trying to weigh it all as much as I used to. I just want 1 or 2 things going my way and nothing going against me.

I actually picked that up style betting basketball. When I first started betting basketball, I knew I couldn't weigh all the contradictory factors in my spreadsheet well. So I only bet when I had 2 things going for me and nothing going against me and it worked really well.

I understand what you are saying, but every time I try to weed out the bad plays, I reduce the data set and it usually turns out poorly.

I'll give you one example...I have a play in maiden races where the top early speed rating is the play at certain distances at 4 to 1 and above. It had a great ROI at all odds ranges and cashed about 15% of the time. I decided to look further into it...I saw that horses with less than 2 Quirin Speed Points were terrible, as were those with 8. Eliminating these types on one data set made the ROI HUGE, and it tested nicely on a few other data sets.

What happened? As soon as I eliminated those QSP horses, the next 3 big winners all fell in that rule, and it was enough to make them positive ROI plays. Many times, you are right with the original, and when trying to make it better you are looking at factors that aren't relevant.

This is just one example and it happened more than a few times, luckily not with real money. I learned my lesson after that one burned me. I can't prove this, but my belief is that the very extra data you seek out is what is providing the value on the horses I find in the first place.

[PaceAdvantage]

I looked at a play earlier this year that only had a 10% hit rate. The play required that you wait for an odd of ten to one or higherThat's nothing...

I've developed six spot plays that I have been testing with base $2 bets live since January 1.

Here are my stats thus far through yesterday (Jan 1 - Jan 15):


# of Bets 218
Win % 4.59%
$1 ROI +.83
Wagered $724
Payoff $1328.20
Profit/Loss +604.20
Number of winners 10
Avg Mutuel ($2) $67.11

Payoffs:
27.40
127.6
45.4
59.5
56.8
91.8
31.6
61.4
77.8
91.8

Note that is # of Bets, not # of Races. I can and will bet more than one horse per race. And this is not a flat bet system. I will wager more than one unit on a horse if he qualifies on more than one spot play (I have seen some qualify on all six, so that would be a six unit bet).

Obviously, this is an extreme longshot system...needless to say, I will continue with this method for at least another 200 bets... :lol:

I have zero problem living with a low hit rate if the method is profitable.

[thaskalos]

That's nothing...

I've developed six spot plays that I have been testing with base $2 bets live since January 1.

Here are my stats thus far through yesterday (Jan 1 - Jan 15):


# of Bets 218
Win % 4.59%
$1 ROI +.83
Wagered $724
Payoff $1328.20
Profit/Loss +604.20
Number of winners 10
Avg Mutuel ($2) $67.11

Payoffs:
27.40
127.6
45.4
59.5
56.8
91.8
31.6
61.4
77.8
91.8

Note that is # of Bets, not # of Races. I can and will bet more than one horse per race. And this is not a flat bet system. I will wager more than one unit on a horse if he qualifies on more than one spot play (I have seen some qualify on all six, so that would be a six unit bet).

Obviously, this is an extreme longshot system...needless to say, I will continue with this method for at least another 200 bets... :lol:
If you decide that you need outside funding in order to expand your operation...please promise me that I will be the first person you will think of to call, or PM.

I promise to be a very silent partner...

[PaceAdvantage]

I've learned over the last 24+ years to not get very excited with preliminary results, especially when dealing with longshots...

The key factor for me thus far has not been deciding which types of horses to bet (although that is of course important), but which types of RACES to bet...

[thaskalos]

I've learned over the last 24+ years to not get very excited with preliminary results, especially when dealing with longshots...

The key factor for me thus far has not been deciding which types of horses to bet (although that is of course important), but which types of RACES to bet...
The more original the better I say...

Longshot betting is preferable, IMO...because not many people have the internal fortitude to follow suit.

It takes "balls" (and a big bankroll) to deal with these types of losing streaks.

[gm10]

The article is about gambler in the UK. There is world of difference in the way horses are bet there and here. There you have price certainty when you make a bet. Things get a lot easier when you know exactly what the return is when you make your wager. That goes double when your profit margin is in the single digits.

Plus the final odds are very rounded due to breakage. The difference between 0.95 and evens is 5% - not insignificant.

I have to agree with an earlier poster. The fact that he bought a Jag, then had to sell it, then won it back the same day ... that's a bit iffy.

Still, he says some sensible things.

[JohnGalt1]

When something simple works & you strive to make it better, one usually makes it worse.

That's true for me.

I can be cruising along winning and making a profit, then I decide --- maybe after reading an article or a handicapping book --- to add a few factors to my equations so I can win even more races and money. Then everything goes haywire. My horses still run competitively, some even win, but I get a lot of seconds or first and thirds, or just misses.

I'm back to a more simple approach, at least until I get another "brilliant" idea to add more factors.

[classhandicapper]

I understand what you are saying, but every time I try to weed out the bad plays, I reduce the data set and it usually turns out poorly.

I'll give you one example...I have a play in maiden races where the top early speed rating is the play at certain distances at 4 to 1 and above. It had a great ROI at all odds ranges and cashed about 15% of the time. I decided to look further into it...I saw that horses with less than 2 Quirin Speed Points were terrible, as were those with 8. Eliminating these types on one data set made the ROI HUGE, and it tested nicely on a few other data sets.

What happened? As soon as I eliminated those QSP horses, the next 3 big winners all fell in that rule, and it was enough to make them positive ROI plays. Many times, you are right with the original, and when trying to make it better you are looking at factors that aren't relevant.

This is just one example and it happened more than a few times, luckily not with real money. I learned my lesson after that one burned me. I can't prove this, but my belief is that the very extra data you seek out is what is providing the value on the horses I find in the first place.


I hear you. I agree that you have to be careful about eliminating horses within a profitable play.

The case I am talking about would be one where you had your maiden play, but inside the same race you also had another automatic play. So two horses qualified for plays. I'd be less apt to play that race (or I'd box them).

I'd prefer playing a race where I had your maiden play and I also had a major negative play on another horse or where the same horse qualified for two different plays. Then I'd have two stars lined up an no reason to have weigh negatives vs. positives for the same horse or positives for two different horses.

[windoor]

That's nothing...

I've developed six spot plays that I have been testing with base $2 bets live since January 1.

Here are my stats thus far through yesterday (Jan 1 - Jan 15):


# of Bets 218
Win % 4.59%
$1 ROI +.83
Wagered $724
Payoff $1328.20
Profit/Loss +604.20
Number of winners 10
Avg Mutuel ($2) $67.11

Payoffs:
27.40
127.6
45.4
59.5
56.8
91.8
31.6
61.4
77.8
91.8

Note that is # of Bets, not # of Races. I can and will bet more than one horse per race. And this is not a flat bet system. I will wager more than one unit on a horse if he qualifies on more than one spot play (I have seen some qualify on all six, so that would be a six unit bet).

Obviously, this is an extreme longshot system...needless to say, I will continue with this method for at least another 200 bets... :lol:

I have zero problem living with a low hit rate if the method is profitable.


Wow, I am impressed.

It is this kind of thinking that can lead to profits. Do what others would normally avoid. Play by the quarter, not by the day, week, or even the month.

Best of luck with it.

Maybe I should dig that play up and revisit it.

Regards,

Windoor

[PaceAdvantage]

That's nothing...

I've developed six spot plays that I have been testing with base $2 bets live since January 1.

Here are my stats thus far through yesterday (Jan 1 - Jan 15):


# of Bets 218
Win % 4.59%
$1 ROI +.83
Wagered $724
Payoff $1328.20
Profit/Loss +604.20
Number of winners 10
Avg Mutuel ($2) $67.11

Payoffs:
27.40
127.6
45.4
59.5
56.8
91.8
31.6
61.4
77.8
91.8

Note that is # of Bets, not # of Races. I can and will bet more than one horse per race. And this is not a flat bet system. I will wager more than one unit on a horse if he qualifies on more than one spot play (I have seen some qualify on all six, so that would be a six unit bet).

Just an update on my progress...now have 504 plays...still showing a healthy profit, but much lower than before...note that I did actually MISS betting on two $40+ winners that would have netted me around an additional $200 in profit (would have had $4 to win on both), but thems are the breaks when you're betting longshots and can NEVER really afford to miss a winner...

These are my stats from Jan 1 - Feb 3 (again, these are 504 bets, not 504 races):

# of Bets 504
Win % 3.57%
$1 ROI +.31
Wagered $1752
Payoff $2295.60
Profit/Loss +543.60
Number of winners 18
Avg Mutuel ($2) $67.80

Payoffs:
27.40
127.6
45.4
59.5
56.8
91.8
31.6
61.4
77.8
91.8
71.4
44.2
67.4
39.0
80.2
45.6
128.6
73

[PaceAdvantage]

And just today I missed an $80+ winner that I would have had $4 on...which is amazing because I have a program that I can type my future bets into and it will monitor the pools and make my bets for me if the odds are high enough...and yet I STILL miss betting on races that I should have won...why?

Because of the way I wrote my program (with my rudimentary programming skills), I can only effectively enter 2 or 3 races at a time into the program...so when life calls, and I get distracted and forget I have another race or two to bet, it is of course THEN that the $80 horse wins without me...

I've also forgotten to bet on a lot of losing races this way as well, but nowhere near enough to make up for the three winners I've missed since I started monitoring this system back on Jan 1...

Bottom line, when betting a super-duper longshot system, you CAN'T MISS a winner...you just can't...it will kill your ROI in the short term dramatically...

[lamboguy]

that system looks great.

5% winners with average payoff of $67 is unreal.

very possible to achieve though when you take out the human emotion of making the picks.

[Robert Fischer]

Money graph:

http://graphics.thomsonreuters.com/11/12/GLB_LPPRCLM1211_VF.jpg

In this graph the problem seems to be more about overconfidence than info-overload.

[aaron]

This is an excellent thread.I tend to lean toward,if its not broken don't fix it. When something is working and showing positive results, I find it best to leave it alone.
For instance,Ed Bain uses trainer stats,with filters to make his bets and he sticks to those stats as long as they are working. He reviews them,but as long as they are working there is no need to change.
One of my problems when handicapping is sometimes,I tend to look at too many things. All this does is take you off bets you should make.