Quote: DJTeddyBearIn case anyone is confused, I took the Yes on those bets. Mike is on the No. And we do the no vig middle.Quote: Wizard
Let the record show I booked the following bets by DJTeddyBear.
link to original post
Thanks for the action again Mike!
link to original post
I agree with Mike that on the bets we make, the "No" is more likely the winning bet.Quote: WizardI should be thanking you, as I think I have the better side of these bets.Quote: DJTeddyBearFrankly, I’m grateful for our friendship as well as the action you give me. 👍
link to original post
link to original post
But don't the odds kinda even things out? Is betting the "Yes" really that bad?
Is the "No", as Mike suggests, the 'better side'? I mean, if they paid even money, then absolutely, but after considering the odds, are the "No" bets really that much better?
My wife also questioned things, saying that I was risking a lot of money. I said yeah, but I only need two of them to come in to be profitable. Hell, if any one of the three big bets come in, I'm making money.
For what it's worth, if I lose every bet, I might re-think things before next year. Otherwise, it's unlikely that your responses will change things.
Side note: Years ago, when I first published our bets, someone responded that taking the average is not the correct mathematical way to calulate a fair middle. I had never thought of that, but it sounds legit.
I'm not sure if that meant it favored me or Mike. Whichever way it was, I assume it would only be a couple bucks different. Why go thru the hastle of the math for that? What's a couple bucks between friends?
I'm certain that it's NOT those couple bucks Mike was referring to when he said he was getting the better side of this.
But in the middle of both curves there's a high spot for yield, things like the BB+ to BBB+ bonds, or the 8-1 horses that aren't attractive to anyone using emotion in their betting/investing but because they're not attracting the emotional betting they have better returns, and if you're an analytical person there are a lot of non-obvious distinctions between the investments in that class that you can work with, and do some sorting.
re horse racing -
faves are the best value when considering ALL of the bets
from the link which I know to be true from other studies:
"long term betting records on Australian horse racing shows a perfectly correlated pattern. Favorites lose the least second favorites lose the 2nd least and so on. The rate of return decreases as price increases
not only does the favorite longshot bias occur in horse racing but also in sports betting"
in horse racing so many dislike the short payout of faves that they become a better value overall - but it would take excellent handicapping to become a long term winner by playing faves - a great many races would be unplayable
the favorite/longshot bias in horse racing has been well known for a great many years
and of course this doesn't mean that a person can't win playing longshots. great selectivity is required. the various studies are about playing ALL longshots
https://www.championbets.com.au/betting-academy-article/short-priced-favourites-can-great-value
.
seems to me,
to take the 'no' is to take the low variance end of the bet, almost always it is the short end btw
to take the 'yes' is to take the high variance end
if it is a fair bet, in the long run it won't matter which end you take. If the long end pays a massive amount, the long run is very, very long
the Wizard has said in articles he believes the squares push the line towards the 'no' on these bets. I think that's why he said what he did
Pretty sure it why the lottery is a thing otherwise if people understood the odds they wouldn’t play it.
Quote: heatmapA real gambler knows the saying “you have to be in it to win it” because even though your a long shot being part of the people who are able to win is enough for most.
Pretty sure it why the lottery is a thing otherwise if people understood the odds they wouldn’t play it.
link to original post
I think most who play the lottery know that the odds are horrible
but for a few bucks they get a pocketful of dreams
.
Quote: lilredrooster.
re horse racing -
faves are the best value when considering ALL of the bets
from the link which I know to be true from other studies:
"long term betting records on Australian horse racing shows a perfectly correlated pattern. Favorites lose the least second favorites lose the 2nd least and so on. The rate of return decreases as price increases
not only does the favorite longshot bias occur in horse racing but also in sports betting"
in horse racing so many dislike the short payout of faves that they become a better value overall - but it would take excellent handicapping to become a long term winner by playing faves - a great many races would be unplayable
the favorite/longshot bias in horse racing has been well known for a great many years
and of course this doesn't mean that a person can't win playing longshots. great selectivity is required. the various studies are about playing ALL longshots
https://www.championbets.com.au/betting-academy-article/short-priced-favourites-can-great-value
.
link to original post
Now I was aware of that study, but was that before or after some wiseguy goes out there telling everyone "The favorites are the best value bets?" No way to tell the actual value of a parimutuel bet before it's run, to be precise.
I'm thinking of it in terms of morning line odds, which aren't worth much but they can tell you which are the favorites and the longshots, and which are likely to be neither.
3 very great books on horse racing although they're very old -
imho the methods of evaluating and comparing the abilities of horses hasn't changed much
they don't involve analyzing tote data though which you expressed interest in in another thread
Tom Ainslie - "Complete Guide to Thoroughbred Racing"
Andrew Beyer - "Beyer on Speed" and "Picking Winners"
Beyer analyzed a horse's speed in a way that was superior to what was in the track program
his speed figures can no longer deliver an edge - but there is much other useful stuff in the books
Beyer crushed racing for a while with his figures using longshots in his exotic bets - he hit a double triple for $190,000 and another one for $134,000
towards the end of his career he complained that due to various reasons (particularly cheating with the use of illegal drugs) it was much harder to beat racing
towards the end of his career he said that with simulcasting you can still find 5 or 6 overlays per day - but you might have to analyze 50 races to find them
the link is a pretty good bio of the man - good luck
https://www.cigaraficionado.com/article/the-andycapper-7603
.
Consider a bet where the bettor has to lay 1000-to-1 compared to one where she is getting 1000-to-1 with an estimated probability of 99.9% that the favorite wins. Neither bet has an edge, but what happens when the probability changes very slightly? When betting the favorite, the edge can never be higher than 0.1% and that only happens when the probability approaches 100%. When betting the long odds, if the probability increased from 0.1% to 0.2% the edge would increase from 0 to 100%.
The people offering the bets understand this and will usually put much more vig on the longshot than the favorite. It's our job to decide when they've attached so much more vig on the longshot that the favorite end up with a negative house edge.
Quote: DJTeddyBearSide note: Years ago, when I first published our bets, someone responded that taking the average is not the correct mathematical way to calulate a fair middle. I had never thought of that, but it sounds legit.
If there is equal vig on both sides of the bet, then the mid-point becomes a bigger and bigger edge as the odds become longer and longer. But it's doubtful there is equal vig on both sides.
Quote: TomGA good bet at long odds is a lot better than a good bet at short odds.
Consider a bet where the bettor has to lay 1000-to-1 compared to one where she is getting 1000-to-1 with an estimated probability of 99.9% that the favorite wins. Neither bet has an edge, but what happens when the probability changes very slightly? When betting the favorite, the edge can never be higher than 0.1% and that only happens when the probability approaches 100%. When betting the long odds, if the probability increased from 0.1% to 0.2% the edge would increase from 0 to 100%.
T
Not really.
The expected logarithmic growth rate is essentially identical in both cases at 0.1%.
For the favorite at p=1:
Edge = 0.1%
Kelly fraction = 100% of bankroll
Expected growth per bet = ln(1 + 1×0.001) = 0.001 or 0.1%
For the longshot at p=0.002:
Edge = 100%
Kelly fraction = 0.1% of bankroll
Expected growth per bet = ln(1 + 0.001×1) = 0.001 or 0.1%
Quote: Archvaldor1
Not really.
The expected logarithmic growth rate is essentially identical in both cases at 0.1%.
For the favorite at p=1:
Edge = 0.1%
Kelly fraction = 100% of bankroll
Expected growth per bet = ln(1 + 1×0.001) = 0.001 or 0.1%
For the longshot at p=0.002:
Edge = 100%
Kelly fraction = 0.1% of bankroll
Expected growth per bet = ln(1 + 0.001×1) = 0.001 or 0.1%
-the chance the favorite wins is capped at 100%; sometimes a longshot can have a chance of winning that is greater than 0.2%.
-this doesn't account for the opportunity cost. If it takes more than a day or it's on a busy day, we'll be worse off getting a 0.1% return on our bets because of everything we'll be missing with higher edges.
-in real world scenarios, very few sportsbooks allow us to bet full Kelly and that's only on the largest markets. If we click max bet, the longshot will be a much higher Kelly multiplier than the favorite, which should give us higher EG.
Logarithmic growth rate, eh?Quote: Archvaldor1Quote: TomGA good bet at long odds is a lot better than a good bet at short odds.
Consider a bet where the bettor has to lay 1000-to-1 compared to one where she is getting 1000-to-1 with an estimated probability of 99.9% that the favorite wins. Neither bet has an edge, but what happens when the probability changes very slightly? When betting the favorite, the edge can never be higher than 0.1% and that only happens when the probability approaches 100%. When betting the long odds, if the probability increased from 0.1% to 0.2% the edge would increase from 0 to 100%.
T
Not really.
The expected logarithmic growth rate is essentially identical in both cases at 0.1%.
For the favorite at p=1:
Edge = 0.1%
Kelly fraction = 100% of bankroll
Expected growth per bet = ln(1 + 1×0.001) = 0.001 or 0.1%
For the longshot at p=0.002:
Edge = 100%
Kelly fraction = 0.1% of bankroll
Expected growth per bet = ln(1 + 0.001×1) = 0.001 or 0.1%
link to original post
I'm not qualified to criticize what you wrote, so I won't call it a bunch of mumbo jumbo
Quote: TomG-in real world scenarios, very few sportsbooks allow us to bet full Kelly
in real world horse racing you can bet as much as you want
you might often find a seeming lock fave at 2/5 in a five horse field betting to show (so many U.S. tracks have small fields now)
minimum payout is $2.10 which it almost always is - 5% profit
it might even be a minus pool meaning there is a shortfall between what is left in the net pool to be distributed to the winning tickets which a payout of the guaranteed minimum (2.10 for every $2.00 bet at most tracks) is mandatory
to me, the bet is very unattractive even if I thought I had an edge
risk $1K to profit by $50 -
no thanks
the horseplayers who do make this kind of bet are called "bridge jumpers" for obvious reasons
.
Quote: lilredrooster
risk $1K to profit by $50 -
no thanks
And this is why you are not an AP.
If you have identified a specific situation where a horse will finish in the top three 98% of the time, but can get 5% return, you have to be a moron not to bet SOMETHING on it. Exactly how much is multi factorial. I’m pretty risk averse, but would probably go into the 6 figures with such an opportunity. Lose $100k 1 out of 50, win $5k 49 out of 50. Sign me up.
Quote: SOOPOOQuote: lilredrooster
risk $1K to profit by $50 -
no thanks
and this is why you are not an AP
If you have identified a specific situation where a horse will finish in the top three 98% of the time, but can get 5% return, you have to be a moron not to bet something on it
you can never know that a horse will finish in the top three 98% of the time
it is only the rawest kind of estimate
there is rampant cheating in horse racing - it's well known - illegal drugs are given horses and there is other kinds of cheating
especially if it's a small track you can't know that the race won't be fixed
you can't know that a jock at a small track who doesn't make that much won't be paid big bucks to stiff his horse by some plunger betting on the others to show
it happens
I might think I have an edge bridge jumping - but I can't know it for sure - so once again - no thanks
as far as me not being an AP - I'm not going to claim I beat racing - I can only say that I lost considerably less than the takeout - my losses were smallish
but in my day I made a lot, lot more in blackjack than you are making betting sports
.
Quote: SOOPOOQuote: lilredrooster
risk $1K to profit by $50 -
no thanks
If you have identified a specific situation where a horse will finish in the top three 98% of the time, but can get 5% return, you have to be a moron not to bet SOMETHING on it. Exactly how much is multi factorial. I’m pretty risk averse, but would probably go into the 6 figures with such an opportunity. Lose $100k 1 out of 50, win $5k 49 out of 50. Sign me up.
link to original post
Some states have a minimum 10% return. I like that better.
But it is a rough way to make a profit given the risk of catastrophic failure for a horse. It doesn't matter how fast he is if there's a stumble or some other problem coming out of the gate, something happens to the jockey, or a disqualification. Even if it's a field of 4. This happened to me once betting dogs, and I had bet a strong favorite, he got bit by another dog and trotted the rest of the race in protest. I'm sure there is data like this out there, how often a strong morning favorite DNF, or is out of the money in a given field size, and Kelly bet sizing would have to apply to that just like anything else.
Quote: AutomaticMonkeySome states have a minimum 10% return. I like that better.
those tracks are allowed to remove show wagering from their menu when they smell the likelihood of bridge jumpers
so the jumpers have to bet Place - which carries more risk
.
Quote: odiousgambit
Logarithmic growth rate, eh?
I'm not qualified to criticize what you wrote, so I won't call it a bunch of mumbo jumbo
That means how much money you make if you take the risk into consideration.
What I know about logarithmic representation of data is that it sometimes makes a better graph, as in where growth has been exponential or compounding. You see this with long term stock market graphs, for instanceQuote: Archvaldor1Quote: odiousgambit
Logarithmic growth rate, eh?
I'm not qualified to criticize what you wrote, so I won't call it a bunch of mumbo jumbo
That means how much money you make if you take the risk into consideration.
link to original post
As for making money exponentially or gradually, doing it by looking at betting on the short end of the bet, or the long end of the bet, you aren't going to find +EV doing that and won't get growth rates at all. Or did I get lost in this discussion somewhere?
Quote: odiousgambitWhat I know about logarithmic representation of data is that it sometimes makes a better graph, as in where growth has been exponential or compounding. You see this with long term stock market graphs, for instanceQuote: Archvaldor1Quote: odiousgambit
Logarithmic growth rate, eh?
I'm not qualified to criticize what you wrote, so I won't call it a bunch of mumbo jumbo
That means how much money you make if you take the risk into consideration.
link to original post
As for making money exponentially or gradually, doing it by looking at betting on the short end of the bet, or the long end of the bet, you aren't going to find +EV doing that and won't get growth rates at all. Or did I get lost in this discussion somewhere?
link to original post
The point was simply that with shorter odds you have lower risk. So you can bet more with the same edge than you could odd with longer-odds wagers.
Each month they give away a $250 betting voucher that must be bet on a single race. It can be split into any number of bets, but the whole $250 on one race.
When a non-horse person wins it, I advise them to put $50 on the favorite, $40 on the next favorite, $30 on the next, $25, $20, $15, etc. Maybe adjust the numbers depending on how many horses are running. Then maybe a few bucks left over to throw at some exactas or trifectas.
In other words, in horse racing, there are no two option bets. Hell, even if you like a particular horse, now you gotta decide if you wanna bet it to come in 1st, 2nd or 3rd.
Unless I'm missing something, and ignoring parlay bets, sports bets are almost all two options per bet. Yes/no or over/under.
Over/under bets get the line specified (and sometimes moved) so there isn't much difference between the two sides. But the difference between the favorite and long shot in the yes/no bets is more obvious and often much more pronounced.
So that leads me to the point of this whole thread and today’s followup.
In Mike's Superbowl 59 results post and the follow-up post, he commented that he avoided the big bridge jumper bets that he's made in the past, primarily because of losing them several times recently.
That leads me to believe that I'm right: The long shot bets aren't quite the bad bet that the 'sharp' bettors make them out to be.
Am I right?
uh oh.Quote: DJTeddyBearIn Mike's Superbowl 59 results post and the follow-up post, he commented that he avoided the big bridge jumper bets that he's made in the past, primarily because of losing them several times recently.
That leads me to believe that I'm right: The long shot bets aren't quite the bad bet that the 'sharp' bettors make them out to be.
Am I right?
link to original post
I've been questioning about what the Wizard's affection for the "no" is on proposition bets and not wanting to bring it up. Now you've broken the ice. If he gets mad it's on you !!
Ha ha. All I've gotten out of it is that the W, has confirmed that the HE is lower, generally, on the 'no'. However, he hasn't said it's +EV. So I think I'd like to hear DJTbear would like to hear why there seems to be this enthusiasm to bet large amounts of money on proposition bets.
link to original post
linked are 2 academic studies - one of college basketball and one of college basketball and college football
6 different authors collaborated on these studies - 3 on each one
both concur that there is a favorite/longshot bias and that the favorite is the better bet
the 2nd link states that betting on a heavy fave is a break even proposition overcoming the house edge
it is possible that it's different in pro sports - I wasn't able to find an academic study of pro sports but I tend to doubt that it is any different
my personal belief is that there is a favorite/longshot bias in all sports and that overall the fave is the better bet
which of course doesn't mean it's necessarily a better bet if you are handicapping - but something to be aware of
https://www.sciencedirect.com/science/article/abs/pii/S1544612304000273
https://www.sciencedirect.com/science/article/abs/pii/S1062976916000041
.
.
But the favorites are also referred to as where the smart money goes. That would imply that the long shots are where the dumb bets are. That's the part that I am questioning. What's so dumb about it?
Maybe the favorite IS the better bet. But maybe it's better the same way that betting the don't pass has a slightly lower house edge than the pass.
Quote: odiousgambituh oh.Quote: DJTeddyBearIn Mike's Superbowl 59 results post and the follow-up post, he commented that he avoided the big bridge jumper bets that he's made in the past, primarily because of losing them several times recently.
That leads me to believe that I'm right: The long shot bets aren't quite the bad bet that the 'sharp' bettors make them out to be.
Am I right?
link to original post
I've been questioning about what the Wizard's affection for the "no" is on proposition bets and not wanting to bring it up. Now you've broken the ice. If he gets mad it's on you !!
Ha ha. All I've gotten out of it is that the W, has confirmed that the HE is lower, generally, on the 'no'. However, he hasn't said it's +EV. So I think I'd like to hear DJTbear would like to hear why there seems to be this enthusiasm to bet large amounts of money on proposition bets.
link to original post
link to original post
It is just common sense. People generally like to bet on things happening, not things not happening.
With regard to favourite betting, yes there are multiple studies showing favourite-longshot bias, not even in just American sports,. People like a big payout-bookies compensate by skewing value towards favourites to balance their books.
This does NOT mean sharps bet favourites. This is because favourite bets don't have much inherent volatility so when they get it wrong the smart gambler doesn't have that much advantage. The best opportunities tend to occur in the middle of the range.
excellent handicapping can get you an edge
knowing about the favorite/longshot bias may have some value - but not very much
Andrew Beyer in his best years crushed racing with longshots
he wouldn't even have dreamed about making a big money bet on a fave bet way down
.
I should address thisQuote: DJTeddyBearFYI: In no way am I implying that I think Mike's bets are dumb. Quite the contrary. For a math guy, those are certainly the correct bets to make.
No, you didn't imply that, I agree.
you've been asking this all alongQuote:But the favorites are also referred to as where the smart money goes. That would imply that the long shots are where the dumb bets are. That's the part that I am questioning. What's so dumb about it?
I was trying to inject some humor into this in the way I did it, and I suppose I was putting words in your mouth. Just taking a little license there . SorryQuote:Maybe the favorite IS the better bet. But maybe it's better the same way that betting the don't pass has a slightly lower house edge than the pass.
link to original post
I think it's pretty clear that the most that can be said for 'taking the no' on proposition bets is two things,
*you usually win the bet, in less than even money payoff , note well, nonetheless we can say the variance is lower ...
*and I think I can say that the Wizard has established that you can expect HE is also lower. Not just thinks... knows
To me, though, seems like the best idea is to stay away from proposition bets, because I am pretty sure the HE is higher generally. That doesn't mean you can't make them for fun of course.
I wish I hadn't said anything, the Wizard is not engaging and now DJTBear is also mad at me
I’m NOT mad. I just thought I needed to clarify.Quote: odiousgambit… I wish I hadn't said anything, the Wizard is not engaging and now DJTBear is also mad at me
link to original post
And, yeah, I kinda wished Mike would have thrown in his two cents…