CFB51 College Football Fan Community
The Power Five => Big Ten => Topic started by: Cincydawg on April 06, 2018, 04:54:45 PM
-
http://www.bettingtalk.com/win-probability-percentage-point-spread-nfl-nba/
This topic has always interested me, and the above is just one version of associating how often a favorite wins relative to the line. For example, a 3 point favorite will win almost 60% of the time, which is pretty close to a coin flip. At 7 points, it's 75%, and at 10 is 84%, and 14 is 92%. We usually think of a 2 TD favorite as being almost prohibitive, akin to a game of Indiana playing at Ohio State, etc. But the dog has an 8% chance of winning, and with as many games as are played each week, there is a good chance some 2 TD dog will win, and if it is a major game we notice.
This site goes to 100% at 17 points, which of course is not correct, but 20+ point favorites that win are major upsets, and we had Stanford beating USC a decade or so back as a 42 point dog. That probably is a 1 in 10,000 shot.
If I think about this over the course of the season for a team like say Ohio State, presume they are favored in every game from 3 points in one game to 30 points in others, but also 7 points in a couple and 10 points in a couple. It's not hard statistically to see a probable loss there somewhere, which is why undefeated teams are rare, even with a light schedule.
A way I try and calculate each team's likely end of year record is to take these percentages and multiply them out, chances of losing, which is a bit of guesswork, and see how many losses are "probable". This usually generates more losses than one would think, or that really happen, over the year. The above scenario for example cranks out 2.4 losses for Ohio State in this imaginary example. I'm curious if anyone can spot an error in this methdology, aside from the guesswork's being wrong.
-
https://www.beyondtheboxscore.com/2009/8/29/1003957/chance-of-a-football-team-winning
I like this link a bit better as it has graphics and more realistic odds I think.
-
You'd be best served becoming an expert in roulette. I know I read that somewhere.
-
You'd be best served becoming an expert in roulette. I know I read that somewhere.
I don't know about that... I think there's a legitimate ground for skill to give you an edge here.
Years ago, I spent an entire season publicly posting and tracking picks on my blog. Over the course of one season (2005 or 2006, can't remember which), I was picking 62.5% ATS, and all you need to beat the vig is 55%. I think college has much more variance (and probably a lot less attention) than picking NFL, and it's a lot easier to spot places where the line is just flat out wrong.
I did this through several ways:
- I did not pick games involving Purdue or Notre Dame. Too much emotional bias.
- I did not pick weeks 1 & 2 because it's too early to know what teams look like due to the degree of turnover on college rosters.
- I primarily picked Big Ten games because those were the programs with which I had a high degree of familiarity.
- Outside the Big Ten, I looked for games where I had a gut feel that the spread might be "wrong", and if my additional research backed up that claim, I included it in my spread tracking.
Essentially what I found was that I'd be picking about 6-8 games per week. Assuming 8 games per week and betting $110 per game (to win $100, just using this for the math), I'd average winning 5 bets per week and losing 3. So I'd gain $500 per week vs a loss of $330, or $170 per week. Averaged over a season, I could've pulled down a good $1500 or so.
The follow-up question, of course, is "well then why don't you do it?"
To that I say this... It takes a lot of time and effort to put into those picks. At the time I didn't really want to devote $800 a week to a betting pool, knowing that a bad week could hit me hard. And all that work and research over a span of 3-4 months, for only $1500? It just wasn't a big enough win to justify the time and risk. If I'd actually wanted to make $1500 over that time, I probably could have found much less risky ways to do so.
To make enough money to be worth it also meant risking a lot more, and especially in 2005/2006 I didn't have a significantly higher bankroll (say betting $550 per game x 8 per week) to do it.
That said, I'd bet that a lot of us on this site could pick games at a money-winning clip. Don't you think you could do better than a 55% record against the spread if you really put your mind to it?
-
:shut_up:
-
My post really is not about betting at all other than using the spread as an input variable. It's more about the nature of upsets, and how "we" seem shocked at times when they happen. We know they WILL happen of course, but not when, but this suggests that they become statistically likely if you have enough games of say 14 point dogs.
-
sh1t happens much more often if I have $$$ on the game
-
FF that seems eerily familiar.About 10 years back on a saturday I put $200 on 1 game and $50 on 3 others.I won the 3 - $50 bets and lost the $200 wager.Go 3 for 4 and lose coin,just seems to be the way it works.It's a real sad story
-
I don't know about that... I think there's a legitimate ground for skill to give you an edge here.
Years ago, I spent an entire season publicly posting and tracking picks on my blog. Over the course of one season (2005 or 2006, can't remember which), I was picking 62.5% ATS, and all you need to beat the vig is 55%. I think college has much more variance (and probably a lot less attention) than picking NFL, and it's a lot easier to spot places where the line is just flat out wrong.
I did this through several ways:
- I did not pick games involving Purdue or Notre Dame. Too much emotional bias.
- I did not pick weeks 1 & 2 because it's too early to know what teams look like due to the degree of turnover on college rosters.
- I primarily picked Big Ten games because those were the programs with which I had a high degree of familiarity.
- Outside the Big Ten, I looked for games where I had a gut feel that the spread might be "wrong", and if my additional research backed up that claim, I included it in my spread tracking.
Essentially what I found was that I'd be picking about 6-8 games per week. Assuming 8 games per week and betting $110 per game (to win $100, just using this for the math), I'd average winning 5 bets per week and losing 3. So I'd gain $500 per week vs a loss of $330, or $170 per week. Averaged over a season, I could've pulled down a good $1500 or so.
The follow-up question, of course, is "well then why don't you do it?"
To that I say this... It takes a lot of time and effort to put into those picks. At the time I didn't really want to devote $800 a week to a betting pool, knowing that a bad week could hit me hard. And all that work and research over a span of 3-4 months, for only $1500? It just wasn't a big enough win to justify the time and risk. If I'd actually wanted to make $1500 over that time, I probably could have found much less risky ways to do so.
To make enough money to be worth it also meant risking a lot more, and especially in 2005/2006 I didn't have a significantly higher bankroll (say betting $550 per game x 8 per week) to do it.
That said, I'd bet that a lot of us on this site could pick games at a money-winning clip. Don't you think you could do better than a 55% record against the spread if you really put your mind to it?
Your point in regards to the first 2 weeks is the reason my brother spends the first 2-3 weeks of each season in Vegas. He contends the odds makers don't have a clue until the third or fourth week. 2010 was the first year he stayed more than 2 weeks. You mentioned ND, so I am sure you remember what occurred to them on that third week. My brother will never forget it. Just before the late game between ND and MSU started, he found himself holding a $250 seven game parlay ticket in which he had picked the first six games correctly and he had MSU as the winner of the seventh game. He still proudly carries the $500 losing ticket he placed as a hedge bet on ND. The Sports Book at the casino didn't have enough cash to pay off his $18,750 (75 to 1) winning parlay ticket. He had to take it to the main cashier in the casino.
-
I agree. My own history loves 1st couple weeks. Big arbitrage imo. I never pick more than 3 games per week. I get crushed in bowl games, and not so well in Nov.
-
Bowl games also favor the dog I think in betting. I also think you might beat the vig by betting against the NDs of the world.
Too many bet for their team to win, so find the team that has the most support and bet against them.
But my point is how often upsets happen.
-
Hah, you just keep trying to keep this thread on point with this crew
see how that works
heck, you know how this works
-
Which team has the prettiest cheerleaders? :57:
-
Which team has the prettiest cheerleaders? :57:
You can win that way, like betting colors at the horse track.
-
Is anyone surprised that a 7 point dog wins 30% of the time? One game in three in effect.
And a 14 point dog wins one time in about 15 or so.
Often we don't notice because the game is between ETSU and ECU or something.
What is the most letters in an FBS football team name not counting the "mascot" part? East Tennessee State University?
-
Is anyone surprised that a 7 point dog wins 30% of the time? One game in three in effect.
And a 14 point dog wins one time in about 15 or so.
Often we don't notice because the game is between ETSU and ECU or something.
What is the most letters in an FBS football team name not counting the "mascot" part? East Tennessee State University?
MTSU would have two more letters than ETSU.
-
MTSU would have two more letters than ETSU.
maybe an a&m school. texas a&m's "a&m" doesn't officially stand for anything per their website. alabama a&m is officially the 'alabama agricultural and mechanical university' which is 42. but they are fcs, and maybe a state with a longer name could beat it anyway.
georgia tech has 28. etsu has 28, mtsu has 30. utep has a 28.
ucla has 35.
the citadel has 44 officially, counting the comma. but fcs again.
but i think va tech has it won for fbs. Virginia Polytechnic Institute and State University - 46.
-
Anyway, the next time your team is a 7 point favorite be VERY AFRAID.
I imagine most of us tend to be pessimists anyway.
We had an interesting discussion a decade or so ago with The Bobs about the home field advantage (HFA). The odd thing was that the SEC in conference games had the smallest HFA of any P5 conference (about 3 points). I think "we" often tend to exaggerate the real impact of HFA. I realize playing at Penn State at night is difference from Indiana of course.
I hope Gatorama has a good cable package in Heaven.
-
Is anyone surprised that a 7 point dog wins 30% of the time? One game in three in effect.
And a 14 point dog wins one time in about 15 or so.
no, but only because I've been picking ATS for the past ten or so seasons in a pick 6 pool
-
https://www.beyondtheboxscore.com/2009/8/29/1003957/chance-of-a-football-team-winning
I like this link a bit better as it has graphics and more realistic odds I think.
I used this same research to hone my picks for the "Stupid Upsets Game" a few years ago and it has never really worked out for me. Here was how I used it:
- A one point dog has a 45% chance of winning so, on average, that pick is worth 0.45 in SUG.
- A 3.5 point dog has a 40% chance of winning so, on average, that pick is worth 1.4 in SUG.
- A 5.5 point dog has a 35% chance of winning so, on average, that pick is worth 1.93 in SUG.
- A seven point dog has a 30% chance of winning so, on average, that pick is worth 2.1 in SUG.
- An eight point dog has a 20-25% chance of winning so, on average, that pick is worth 1.6 - 2 in SUG.
- An 11.5 point dog has a 15% chance of winning so, on average, that pick is worth 1.73 in SUG.
- A two TD dog has a 5-10% chance of winning so, on average, that pick is worth 0.7 - 1.4 points in SUG.
Thus, at least in theory, the best SUG picks are in the range of about one TD. That said, this has never worked out for me as I've never won.
-
I hope Gatorama has a good cable package in Heaven.
LittleBigMatt, Mr. Hoople and NUWildcat too.
-
I don't know about that... I think there's a legitimate ground for skill to give you an edge here.
Years ago, I spent an entire season publicly posting and tracking picks on my blog. Over the course of one season (2005 or 2006, can't remember which), I was picking 62.5% ATS, and all you need to beat the vig is 55%. I think college has much more variance (and probably a lot less attention) than picking NFL, and it's a lot easier to spot places where the line is just flat out wrong.
I did this through several ways:
- I did not pick games involving Purdue or Notre Dame. Too much emotional bias.
- I did not pick weeks 1 & 2 because it's too early to know what teams look like due to the degree of turnover on college rosters.
- I primarily picked Big Ten games because those were the programs with which I had a high degree of familiarity.
- Outside the Big Ten, I looked for games where I had a gut feel that the spread might be "wrong", and if my additional research backed up that claim, I included it in my spread tracking.
Essentially what I found was that I'd be picking about 6-8 games per week. Assuming 8 games per week and betting $110 per game (to win $100, just using this for the math), I'd average winning 5 bets per week and losing 3. So I'd gain $500 per week vs a loss of $330, or $170 per week. Averaged over a season, I could've pulled down a good $1500 or so.
The follow-up question, of course, is "well then why don't you do it?"
To that I say this... It takes a lot of time and effort to put into those picks. At the time I didn't really want to devote $800 a week to a betting pool, knowing that a bad week could hit me hard. And all that work and research over a span of 3-4 months, for only $1500? It just wasn't a big enough win to justify the time and risk. If I'd actually wanted to make $1500 over that time, I probably could have found much less risky ways to do so.
To make enough money to be worth it also meant risking a lot more, and especially in 2005/2006 I didn't have a significantly higher bankroll (say betting $550 per game x 8 per week) to do it.
That said, I'd bet that a lot of us on this site could pick games at a money-winning clip. Don't you think you could do better than a 55% record against the spread if you really put your mind to it?
I learned gambling from some guys who lived on my floor freshman year. They were booking bets and I made a few small bets with them and learned from them how it worked. They absorbed WAY too much risk and ended up having to drop out and get jobs to pay everyone back.
What I learned was that it was way to risky for me to want to be a bookie and I'm generally a lot less risk averse than the average bear. The problem is this:
The bookie takes 10% from the losers but for the bookie that only works out to 5% of the total. If you are booking $10k a week in bets that 5% is only $500/week. Worse, it isn't really even that high. Bets tend to be very regional. Ie, these guys were booking bets in a dorm at Ohio State so nearly every week they had more bet on tOSU than against tOSU. Explanation for the initiated:
If I'm a bookie and bwar bets $1,000 on Purdue+10.5 at Wisconsin while badge bets $1,000 on Wisconsin-10.5 vs Purdue this is AWESOME for me. Either way I make $100:
- If Purdue wins or loses by <10.5 I get $1,100 from badge and give $1,000 to bwar.
- If Wisconsin wins by >10.5 I get $1,100 from bwar and give $1,000 to badge.
The problem for me, as the hypothetical bookie, is if badge and bwar bet the same way. Lets say they each bet $1,000 on PU+10.5. Now the possibilities are:
- If Wisconsin wins by >10.5 then I get $1,100 each from bwar and badge and pocket $2,200.
- If Purdue wins or loses by <10.5 then I owe bwar and badge $1,000 each and I'm out $2k.
The only way for me, as the hypothetical bookie, to get out of having effectively bet $2k on UW+10.5 is to pass that bet along by calling a bigger bookie and betting $2k on Purdue+10.5. The problem is that I don't make any money at all that way. Now all am is a pass-through. If Purdue wins or loses by >10.5 then I get $2k from my bookie and give it all to bwar and badge. If Wisconsin wins by >10.5 then I take bwar's and badge's money and give it to my bookie.
Thus, even if I could book $10k/week in bets I'd still have to pass a decent percentage of it on. These two kids were booking bets at Ohio State. They tended to get a bunch of bets for Ohio State from naive young Buckeye fans. If Ohio State didn't cover the spread these two kids made a bunch of cash but if they did, well they weren't smart enough to pass their losers on so they were screwed.
Then, on top of that you have the issue of collections. There is a reason that the mob breaks people's kneecaps if they don't pay. Gambling is illegal. In the US our courts cannot be used to collect money earned through illegal means. Therefore, if either bwar or badge refuse to pay I'm on my own to collect the money and my clients are not going to be interested in rain-checks or "bookie credit". They want cash.
Obviously the collectibility issue is bigger when you take bets from people you don't know well. If you keep your operation small it might not be a problem at all but you can't make any money running a small bookie operation. You could also eliminate collections issues by requiring payment upfront but again, you can't run a large bookie operation with that restriction. If you want to make a decent amount of money at it you have to be booking at least $10k/week. Even then you are going to net less than $500/week by the time you back out for bets that you have to pass on to a bigger bookie and/or collections issues. For that amount of work it just isn't worth it.
-
LittleBigMatt, Mr. Hoople and NUWildcat too.
I figure they are together. Sad list, thanks for posting.
-
medina,
Agreed... Being a small-time bookie is a lot of headache and risk for not so much money. Basically if you're a small-time bookie, you're not only facilitating gambling, you're essentially gambling yourself not only that you won't have too much action go against you and wipe out your ability to repay your winners, but also gambling that you can collect from your losers.
-
I used this same research to hone my picks for the "Stupid Upsets Game" a few years ago and it has never really worked out for me. Here was how I used it:
- A one point dog has a 45% chance of winning so, on average, that pick is worth 0.45 in SUG.
- A 3.5 point dog has a 40% chance of winning so, on average, that pick is worth 1.4 in SUG.
- A 5.5 point dog has a 35% chance of winning so, on average, that pick is worth 1.93 in SUG.
- A seven point dog has a 30% chance of winning so, on average, that pick is worth 2.1 in SUG.
- An eight point dog has a 20-25% chance of winning so, on average, that pick is worth 1.6 - 2 in SUG.
- An 11.5 point dog has a 15% chance of winning so, on average, that pick is worth 1.73 in SUG.
- A two TD dog has a 5-10% chance of winning so, on average, that pick is worth 0.7 - 1.4 points in SUG.
Thus, at least in theory, the best SUG picks are in the range of about one TD. That said, this has never worked out for me as I've never won.
i always look for home dogs, and particularly early in season, non-conf long distance home dogs.
-
Somebody mentioned Roulette and "becoming an expert" at it. In case anyone is unaware, that is funny because Roulette is a game in which there is literally zero advantage to "being an expert". A Roulette wheel has 38 spots and pays such that it would be exactly fair (ie, you would neither make nor lose any money in the long-run) if there were 36. Effectively the two green spots (known as 0 and 00) are the casino's take. It pays 36/38 or 94.7368%. Some bets to illustrate the example:
- If you bet either red or black you will win, on average, 18 out of 38 times and it pays 2:1. If you bet $10 on red 38 times you will, on average, win 18 times. You will spend $380 and win $360 (18*20) thus losing $20 out of $380 which is 5.2632%
- If you bet 1-12, 13-24, or 25-36 you will win, on average, 12 out of 38 times and it pays 3:1. If you bet $10 each of 38 times you will win, on average 12 times. You will spend $380 and win $360 (12*30) thus losing $20 out of $380 which is 5.2632%.
- If you bet a particular number (any of the 38) you will win, on average, once out of 38 times and it pays 36:1. If you bet $10 on a particular number 38 times you will, on average win once. You will spend $380 and win $360 (1*360) thus losing $20 out of $380 which is 5.2632%.
The point is that Roulette isn't like blackjack or craps in which some bets are better than others and you need to know what you are doing in order to maximize your odds. Instead, every Roulette bet has the exact same 94.7368% average payout so it makes absolutely no difference what you bet.
I once read about some guys who had a theory that was almost (but not quite) foolproof to make a relatively small gain by playing Roulette. Here is the theory:
Suppose you have $1,000 and you really need $1,050. You could do this:
- Bet $50 on red on a Roulette table. If you win you'll have the $1,050 and you can walk away up $50. If you lose:
- You now have $950. Bet $100 on red on a Roulette table. If you win you'll have $1,050 and you can walk away up $50. If you lose:
- You now have $850. Bet $200 on red on a Roulette table. If you win you'll have $1,050 and you can walk away up $50. If you lose:
- You now have $650. Bet $500 on red on a Roulette table. If you win you'll have $1,050 and you can walk away up $50. If you lose:
- You now have $150. Bet $112.50 on a corner (four numbers). If you win it pays 9:1 so your $112.50 will become $1,012.50 and, when you add that to the $37.50 that you didn't bet you will now have $1,050 and you can walk away up $50. If you lose:
- You now have $37.50. Bet $28.93 on a particular number. If you win it pays 36:1 so your $28.93 will become $1,041.48 and, you will still have the $8.58 that you didn't bet so you can walk away up $50. If you lose:
- You now have $8.58. There is no bet on a Roulette table that can turn $8.58 into $1,050.
Ignoring #5-7 (because the longer odds are more complicated) it is ALMOST a sure bet than you can "make" 5% on a Roulette table by employing steps 1-4. If you are a little less greedy and you only want to "make" 2.5% the odds are even better (near, but not quite perfect).
From what I remember of statistics (correct me if I am wrong here):
- Chance of losing bet #1 is 52.63% (20 out of 38).
- Chance of losing bet #1 AND bet #2 is 27.70% (.5263*.5263).
- Chance of losing bet #1 AND bet #2 AND bet #3 is 14.579% (.5263^3).
- Chance of losing bet #1 AND bet #2 AND bet #3 AND bet #4 is 7.68% (.5263^4).
Thus, steps #1-4 give you a roughly 94% chance of "making" 5% on your money.
-
medina,
Agreed... Being a small-time bookie is a lot of headache and risk for not so much money. Basically if you're a small-time bookie, you're not only facilitating gambling, you're essentially gambling yourself not only that you won't have too much action go against you and wipe out your ability to repay your winners, but also gambling that you can collect from your losers.
Exactly and, unfortunately for those two guys, they failed to realize it.
Another example that turned me off of gambling:
I don't remember the exact $$$$'s or the date but when I was in school Tyson was still knocking everybody out. He had won a bunch of fights and the pay-per-view customers were starting to complain because they had paid big money for a fight and then ended up only getting a couple rounds or less. This became a major story. It became such a big story that the PPV people announced that, for this particular fight they would cap the PPV fee based on time such that if one of the fighters (Tyson - everyone assumed) knocked out his opponent in the first round the PPV viewers would only have to pay a limited charge.
I might be wrong, but I've LONG believed that you could make money betting boxing simply by figuring out which result was best for Don King and simply betting on that. When I heard these news stories about PPV backlash and whatnot I figured Tyson was about to lose. Tyson was something like a 20:1 favorite in this fight (maybe more, I don't remember). I, as a broke college kid, managed to scrounge up $50 and I wanted to bet my $50 on the underdog. I didn't have a bookie but my roommate did so I asked him to make the bet for me. My roommate told be (before the fight, he didn't take my winnings and screw me) that his bookie wouldn't take it.
I can completely understand the bookie's predicament. In order to balance off my $50 bet on the underdog he would have needed $1,000 worth of bets on Tyson and practically nobody bets favorites that big. What is the fun in betting $200 to win $10? Still, it really ticked me off especially when Tyson lost just like I figured he would.
-
I know a sure fire way to make a small fortune on stock options.
-
I know a sure fire way to make a small fortune on stock options.
I did too, right up until the crash in 2008.
-
I know a sure fire way to make a small fortune on stock options.
Roulette, baby.
-
Buying a winery is one of the usual methods, or buying a restaurant (which I once did, a small part of one).
-
It's like the old question...
Q: Do you know how to make a million dollars in auto racing?
A: Well, first you start with $10 Million, then...
-
I used this same research to hone my picks for the "Stupid Upsets Game" a few years ago and it has never really worked out for me. Here was how I used it:
- A one point dog has a 45% chance of winning so, on average, that pick is worth 0.45 in SUG.
- A 3.5 point dog has a 40% chance of winning so, on average, that pick is worth 1.4 in SUG.
- A 5.5 point dog has a 35% chance of winning so, on average, that pick is worth 1.93 in SUG.
- A seven point dog has a 30% chance of winning so, on average, that pick is worth 2.1 in SUG.
- An eight point dog has a 20-25% chance of winning so, on average, that pick is worth 1.6 - 2 in SUG.
- An 11.5 point dog has a 15% chance of winning so, on average, that pick is worth 1.73 in SUG.
- A two TD dog has a 5-10% chance of winning so, on average, that pick is worth 0.7 - 1.4 points in SUG.
Thus, at least in theory, the best SUG picks are in the range of about one TD. That said, this has never worked out for me as I've never won.
I managed to win one season by selecting between a 10 to 15 point home dog each week. Simple but fairly effective
-
I managed to win one season by selecting between a 10 to 15 point home dog each week. Simple but fairly effective
That is usually roughly how I start out but then about mid-way through the year it becomes apparent that I am too far behind to catch up that way so I start picking truly stupid upset picks.
-
Topics like this always remind me of fundamental misunderstandings about statistics--and I know most of you know these things. That said, even when we know them, the results are often jarring.
As an example: If there's a 30% chance of something happenings, then that thing happening is an expected outcome. 3 out of 10 times is a lot of times. The same is true of a 5% chance; even if its a less frequent outcome, it is still an expected outcome. In sports it's often easier to see the inputs that lead to such an outcome: particularly egregious fouls (or referee's calls), injuries, sub-standard play by key players, etc.
The frequency of the event impacts our reaction to the statistics.
In my profession I see this a lot: "We have a 70% chance of winning at trial." This is a single event, so the reaction is often something like, "Awesome, we're going to win!"
But when we go to the baseball game: "This guy hits .300." There are tens of thousands of at-bats per season, so: "So glad he's at the plate for us!"
Same person reacting to the same odds drawing exactly opposite conclusions. (And that's not even touching how subjective a lot of these odds are.)
And of course, when there is a 90% chance of something happening, but it doesn't happen, that doesn't mean the 90% was wrong.
In politics people often confuse the odds of an event, with the percent of the vote. A 5-point win in a national election (with approximately 130 million voters) is a blowout, but receiving 52.5% of the vote doesn't mean the winner had a 52.5% chance of winning.
And the casino business model is pretty straight forward: it's math. It requires capital to float the math, but with tens of thousands of events (depending on the time period you're measuring), the casino wants you to win because that will convince the other people to come (and lose). While 60% is a coin flip on a single game, 52% is a lot of profit over the course of thousands of events (and ~48% is basically the best odds you'll get in a casino).
So, in the roulette example, assuming the statistics presented earlier were accurate (and it's been a long time since I've done any of that math), with a 94% chance that you win $50, there's a 6 percent chance that you lose $1000. Winning $50 94 times means you've won $4,700. Losing $1000 6 times means you lose $6000. So for every 100 tries at this theory, the casino wins $1300. And that's a theory in which the winners show restraint, and walk away after their 5% payday on $1000. Few people head to the casino to make 5%.
Sports betting has always been a mystery to me because I don't understand how they set the line (I'm sure I could learn it, I just don't care that much), but regardless of the inputs, I'm confident in saying that the casinos know how to do it such that they encourage a lot of winning--just so long as it's a little less than the losing. Your local, college bookie may not have the capital to float the losses, but the big kids do, and they rely on it to keep raking in the dough--and building more capital.
-
I imagine they set the line where they think will elicit an equal number of bets ($$$ wise) on each side. The closer they get, the more stable the line and the happier they will be. It's a sure thing unless the bets get so unbalanced they can't fix it.
With experience, the line gets set because they know ahead of time that OSU fans are numerous and like to bet ON OSU. So, the line when OSU plays IU "should be" 8 points but is set at 10.
I have pondered if one could beat the vig by taking every less popular team in the game. I also think that upsets are more likely in bowl games than in conference games.
But imagine a business where you take bets on a contest and have as much money on one side as the other and you get 10%. Done at scale, you could build huge fancy buildings in the middle of some desert.
The baseball analogy is a good one, a .300 hitter versus a .250 hitter. The odds are not really much different but we tend to think they are. A .400 hitter makes an out 60% of the time (ignoring unofficials) while a .200 does it 80% of the time, both figures being fairly similar in context of a single "try".
-
Sports betting has always been a mystery to me because I don't understand how they set the line (I'm sure I could learn it, I just don't care that much), but regardless of the inputs, I'm confident in saying that the casinos know how to do it such that they encourage a lot of winning--just so long as it's a little less than the losing. Your local, college bookie may not have the capital to float the losses, but the big kids do, and they rely on it to keep raking in the dough--and building more capital.
The casino doesn't care if you win or lose your bet. They want a 50/50 split of the action and they make their cut off the vig. So if they set the line such that an even number of bettors bet on either side of the spread they make money.
That's why I say sports betting might be exploitable in a way that roulette can't. You're not necessarily playing "against the house" in a simple game of chance
Same as with poker The casino takes a cut of every pot, so it's hot against the other players, not against the house.
-
Sports betting has always been a mystery to me because I don't understand how they set the line (I'm sure I could learn it, I just don't care that much), but regardless of the inputs, I'm confident in saying that the casinos know how to do it such that they encourage a lot of winning--just so long as it's a little less than the losing. Your local, college bookie may not have the capital to float the losses, but the big kids do, and they rely on it to keep raking in the dough--and building more capital.
Lines are usually set with mathematical power rankings. I know college basketball basically mirros the KenPom ratings.
-
Talking about collecting money - my core group of friends and I ran a sports pick 'em every week for football season and that was hard enough to get all the money in, accounted for, and paid out.
It was easy when it was just us, but one year it got BIG and there was some serious money fluttering about.
-
That is usually roughly how I start out but then about mid-way through the year it becomes apparent that I am too far behind to catch up that way so I start picking truly stupid upset picks.
that's why you gotta hit a big one early, and then pick more reasonable ones as season goes on.
for the early one, find a west coast team traveling cross country as a big favorite. or vice versa.
-
I suspect a really savvy player can make some dough on sports betting, with a lot of work. As noted above, the House gets their take no matter what, but you are betting against some "fans" out there. I imagine there is some heavy money that comes in if some bet gets out of whack, but perhaps you can still find room between the elephants.
Of course, we had the "sure thing in roulette" champ around here for a bit. Probability didn't seem to be a factor for him.
I'd bet early season games and bowl games and against any team that was "popular".
-
I don't gamble very often but when I do go to a Casino I usually play Roulette. That might surprise some people here because you all know numbers are my thing and Roulette has just about the worst odds in the casino (94.7% payout) but I do have a reason. I don't really plan to win. I see it as an entertainment expense and that is where Roulette has an advantage. Roulette is probably the slowest moving game in the casino. Each time they have to wait for everybody to place their bets then spin the wheel, drop the ball and let it spin until it lands, then they have to pay the winners and start over. It is a LOT slower than blackjack or even Craps. A couple hundred dollars will generally last you for at least several hours on a $10 Roulette table.
Then there is my theory:
If you think about roulette from the casino's perspective, they make most of their money when it hits green. If @Cincydawg (https://www.cfb51.com/index.php?action=profile;u=870) bets red and I bet black the casino gives my money to him when it hits red and his money to me when it hits black. Same thing if we are betting even/odd or whatever but the casino takes both of our money when it lands on green. Consequently, my theory is that if the Casino could slightly increase the odds of landing on green they would. Maybe the green spaces are a millimeter wider than the others or something but generally the casino would make more money that way so I always bet on green.
-
Then there is my theory:
If you think about roulette from the casino's perspective, they make most of their money when it hits green. If @Cincydawg (https://www.cfb51.com/index.php?action=profile;u=870) bets red and I bet black the casino gives my money to him when it hits red and his money to me when it hits black. Same thing if we are betting even/odd or whatever but the casino takes both of our money when it lands on green. Consequently, my theory is that if the Casino could slightly increase the odds of landing on green they would. Maybe the green spaces are a millimeter wider than the others or something but generally the casino would make more money that way so I always bet on green.
I think that would be unlikely, at least in Vegas. The Nevada Gaming Commission is no joke. And the casinos know the fallout with the general public if they were to be proven to "cheat" the game with an unbalanced wheel.
The casinos know they're going to make money even with a perfectly equally weighted wheel. It's built into the game. The risk, both with the gaming commission and with the public's perception, make it highly risky to mess around and not give people the right odds.
Now, you get outside Nevada, and maybe to smaller casinos in some of the less-regulated states or the Indian casinos, and maybe you have a point. But I don't see it happening in Vegas.
-
I think that would be unlikely, at least in Vegas. The Nevada Gaming Commission is no joke. And the casinos know the fallout with the general public if they were to be proven to "cheat" the game with an unbalanced wheel.
The casinos know they're going to make money even with a perfectly equally weighted wheel. It's built into the game. The risk, both with the gaming commission and with the public's perception, make it highly risky to mess around and not give people the right odds.
Now, you get outside Nevada, and maybe to smaller casinos in some of the less-regulated states or the Indian casinos, and maybe you have a point. But I don't see it happening in Vegas.
Yeah, probably not but it is the theory I go with.
-
I suspect a really savvy player can make some dough on sports betting, with a lot of work. As noted above, the House gets their take no matter what, but you are betting against some "fans" out there. I imagine there is some heavy money that comes in if some bet gets out of whack, but perhaps you can still find room between the elephants.
Jimmy the Greek?
-
I suspect a really savvy player can make some dough on sports betting, with a lot of work. As noted above, the House gets their take no matter what, but you are betting against some "fans" out there. I imagine there is some heavy money that comes in if some bet gets out of whack, but perhaps you can still find room between the elephants.
Just for 100% clarity, you are betting against the house, not against other players. That is a key differentiator between this and poker, where you're betting against other players and the house gets their cut.
The house wants 50/50 betting on either side of the line, because when they get that, it doesn't matter which side wins, they make money. If they set the line wrong and it goes 80/20, with the win in favor of the 80% side, the money to fund that comes out of the HOUSE not the 20% who lost their bet (of course they lose to the house). Or to make it simpler, let's say the betting comes in 100% on one side of the line and that side wins. Who pays the winners? The house, because there ARE no losing betters.
But that's why the line can move if there is too much action going one direction. If the betting is coming in lopsided one way or the other, the line will move to entice closer to even action. Because when the action is even, the house CAN'T lose.
My point regarding this, however, is that the "line" may not be right, even when the betting is 50/50. The question is whether a savvy sports bettor can use his/her own SKILL in order to find places where those lines are off.
Because unlike roulette, or blackjack, or video poker, the lines don't reflect an exact mathematical probability of outcome of an impersonal collection of spaces on a wheel, or a defined number of cards, or a random number generator in a CPU. The outcome is not mathematically determined. The outcome is dependent on humans, who are notoriously unpredictable. The question is whether a savvy bettor can determine WHEN the unpredictability leans more heavily a certain way.
With the vig, you don't come out ahead if you're correct until you can predict at a >55% clip. But it's still betting against the house.
-
When I say you are betting against some "fans out there", I mean their attitudes help set the line. If you can discern that too many fans are betting unrealistically you might beat the House take. Maybe.
I realize the bet is with the bookie. But the bookie is trying to get to even bets if he can. If OSU fans like to bet on OSU, you can perhaps go against that popular opinion and get a better line than is really warranted.
-
When I say you are betting against some "fans out there", I mean their attitudes help set the line. If you can discern that too many fans are betting unrealistically you might beat the House take. Maybe.
I realize the bet is with the bookie. But the bookie is trying to get to even bets if he can. If OSU fans like to bet on OSU, you can perhaps go against that popular opinion and get a better line than is really warranted.
Understood. I can be a bit pedantic at times, so I just wanted to clarify in case anyone reading the thread wasn't clear.
-
I realize the bet is with the bookie. But the bookie is trying to get to even bets if he can. If OSU fans like to bet on OSU, you can perhaps go against that popular opinion and get a better line than is really warranted.
I get the theory. In theory OSU, Bama, ND have lots of fans and in theory fans probably prefer to bet on their team so in theory the spread for OSU/Bama/ND should be slanted a little bit such that teams like OSU/Bama/ND with lots of fans should go somewhat below .500 ATS (against the spread).
First, I'd like to see some research. Do teams like tOSU/Bama/ND typically finish below .500 ATS?
My other problem is that in at least half of the games in a typical year a team like tOSU/Bama/ND is obviously substantially better. At that point the spread isn't a casino guess as to how much better tOSU/Bama/ND is, it is a guess as to how much tOSU/Bama/ND will want to win by.
Lets say that Ohio State plays Purdue and the Buckeyes are favored by 17. Mid-way through the 4th quarter the Buckeyes receive a punt and take possession ahead by 23 points. Now I think at that point we could all agree that it sure looks like the Buckeyes are in fact at least 17 points better than Purdue but what happens next has as much to do with what Urban Meyer wants as anything else. If Urban Meyer wants "style points" then there is a decent chance that the Buckeyes will score another TD and win by 30. If Urban Meyer is happy with the performance and the (at that point) sure win and just wants to avoid injuries to starters and/or get a look at the backups in action he'll send out the second and third stringers on offense and when they lose the ball he'll send out the second and third stringers on defense and there is a decent chance that Purdue will score a TD and beat the spread by finishing "only" 16 points down to the 17 point favorite Buckeyes.
I guess part of the reason I've never done much sports betting is that it would frustrate me to no end to watch as the team that I bet on had an easy opportunity to cover and didn't take it because they don't care about covering and just want to win. In my above example it is the same thing if I had bet on Purdue +17 and they were down 19 and in FG range late. Obviously they are going to try for a TD but all I need is a FG.
-
I rarely bet on favorites when the spread is 28 or more - ok, never
cause the coach may pull starters and take the air out of the ball once he gets up by 4 TDs in the 4th quarter
the dog then scores a late TD and poof! my money down the drain
-
http://www.vegasinsider.com/college-football/against-the-spread/
Looks like popular powerhouse teams Ball State, Kansas, Akron, WKU, and UT-San Antonio had bad ATS scores last year.
Hmmmmm ....
-
https://www.teamrankings.com/ncf/trends/ats_trends/
A little easier to read. Iowa State did the best ATS last season.
-
I checked a few other years and my hypothesis is not valid, at all. Derp.
-
I get the theory. In theory OSU, Bama, ND have lots of fans and in theory fans probably prefer to bet on their team so in theory the spread for OSU/Bama/ND should be slanted a little bit such that teams like OSU/Bama/ND with lots of fans should go somewhat below .500 ATS (against the spread).
First, I'd like to see some research. Do teams like tOSU/Bama/ND typically finish below .500 ATS?
I don't think any simple mechanical "bet against helmets ATS" is going to have an edge, because to be honest any simple mechanical system--if it's actually any good--would quickly be exploited to the effect that it would be nil. I think the "efficient market hypothesis" is usually overstated, but along the same lines if Vegas found themselves consistently losing money on helmet teams in the sportsbook, I am ABSOLUTELY sure they would see that and correct.
It's similar to the stock market. Once some systematic investment strategy, such as the "Dogs of the Dow", becomes popular, it gets baked into the investment strategies and then fails to outperform.
-
Nicely put, and completely agree.
-
When finding a bookie was harder, or when lines had more of a spread between bid/ask it was common for the line to be a little juicier in helmet teams home market. Nebraska was 2 or 3 pts more expensive to bet in Lincoln than the posted odds at a sportsbook during my college days. There was a market premium other wise the action was too heavy on the home team .
-
Btw: Bill Snyder and KSU ATS in your portfolio the past 25 yrs was an amazing investment on an annualized basis. Performance hasn't been as strong as bettors started to get wise lately but it's been a Five Star play.
I believe he was 161-111 ATS from 89 to 05. 4-10 in bowls ats.
He was about 66% during his 2nd tour at KSU, though I think a little lower now.
Mike Gundy has also performed high ATS. Rich Rod was a terrible bet iirc.
-
So, no team or coach can beat the spread consistently for very long before the market adjusts.
Bettors aren't dumb. Well, some are, enough to make the others have a shot.
-
I see an ad on CNBC fairly often (I watch while eating lunch occasionally) for some fellow who claims to have a sure fire way to invest in stocks and make huge bucks, and of course he's selling his DVD and book. I figure that a method that really did win BIG, not just modestly, in the markets would soon obsolete itself, really it would happen immediately, and no long work, unless it was kept very quiet and secret known to few.
-
Ergo, if I found some fairly obvious way to beat the spread, well, the "market" would adjust.
-
I see an ad on CNBC fairly often (I watch while eating lunch occasionally) for some fellow who claims to have a sure fire way to invest in stocks and make huge bucks, and of course he's selling his DVD and book. I figure that a method that really did win BIG, not just modestly, in the markets would soon obsolete itself, really it would happen immediately, and no long work, unless it was kept very quiet and secret known to few.
hah, if you can make huge bucks in the stock market, you certainly don't need to sell books and DVDs
-
When finding a bookie was harder, or when lines had more of a spread between bid/ask it was common for the line to be a little juicier in helmet teams home market. Nebraska was 2 or 3 pts more expensive to bet in Lincoln than the posted odds at a sportsbook during my college days. There was a market premium other wise the action was too heavy on the home team .
That might have created an interesting arbitrage opportunity. I'm thinking through this as I type:
Suppose that Nebraska is favored by 6.5 nationally and 9.5 in Lincoln. So in Lincoln I could bet $1,000 on Nebraska's Opponent +9.5 then drive to Chicago and bet $1,000 on Nebraska -6.5. Three possibilities:
- Nebraska loses or wins by <6.5. I win the bet on Nebraska's opponent and lose the bet on Nebraska. I'm out a net $100.
- Nebraska wins by 7-9 points. I win both bets and make $2,000.
- Nebraska wins by >9.5. I win the bet on Nebraska and lose the bet on Nebraska's opponent. I'm out a net $100.
Over the course of a (12 game back then) season I'd only need to hit #2 once. Ie, in a 12 game season I could lose $100 11 times and win $2,000 once and I'd be up $900. If I hit #2 any more than that I'd really make a killing.
-
That might have created an interesting arbitrage opportunity. I'm thinking through this as I type:
Suppose that Nebraska is favored by 6.5 nationally and 9.5 in Lincoln. So in Lincoln I could bet $1,000 on Nebraska's Opponent +9.5 then drive to Chicago and bet $1,000 on Nebraska -6.5. Three possibilities:
- Nebraska loses or wins by <6.5. I win the bet on Nebraska's opponent and lose the bet on Nebraska. I'm out a net $100.
- Nebraska wins by 7-9 points. I win both bets and make $2,000.
- Nebraska wins by >9.5. I win the bet on Nebraska and lose the bet on Nebraska's opponent. I'm out a net $100.
Over the course of a (12 game back then) season I'd only need to hit #2 once. Ie, in a 12 game season I could lose $100 11 times and win $2,000 once and I'd be up $900. If I hit #2 any more than that I'd really make a killing.
Called "hitting the middle". Most famous case was Super Bowl XIII. Called Black Sunday in the bookie business. Line fluctuated between -3.5 and -4.5 with the Steelers favored over the Cowboys. The "middle" bettors gave the 3.5 and took the +4.5. Steelers won by 4 and the bookies took a beating.
http://www.sportingnews.com/nfl/news/super-bowl-2014-odds-line-favorite-broncos-seahawks-betting-las-vegas-1979/14ujqgfsx04sw15iyzfzkbgx9f
-
Called "hitting the middle". Most famous case was Super Bowl XIII. Called Black Sunday in the bookie business. Line fluctuated between -3.5 and -4.5 with the Steelers favored over the Cowboys. The "middle" bettors gave the 3.5 and took the +4.5. Steelers won by 4 and the bookies took a beating.
http://www.sportingnews.com/nfl/news/super-bowl-2014-odds-line-favorite-broncos-seahawks-betting-las-vegas-1979/14ujqgfsx04sw15iyzfzkbgx9f
Thanks for sharing, interesting read.