Tweet
One of the things that absolutely *kills* me, is the number of "handicappers" that go on and on and on about how many "trends" apply to a certain game...making it a near certainty.
Let's get one thing straight from the start....almost ALL of these "trends" aren't "trends" at all....they are minor anomalies....blips on the radar screen that mean NOTHING.
I'll see guys like Mark Lawrence or (god forbid) Brandon Lane (who I commented on at more length in the thread called "update on my friend") raving about how a perfect 6-0 trend applies to a given game. This demonstrates such a lack of sound mathmatical reasoning that it's fucking terrifying that people put their money at risk based on this kind of advice.
First, let's get to the bottom of what actually *causes* most of these *trends* that so many handicappers are so proud to have "discovered".
We all know that if you flip a coin 10x, the odds say that you will get 5 heads and 5 tails. There is NO question that those are the true odds.
BUT....BUT....that doesn't mean that on any given series of 10 flips you WILL get 5 of each. you may well get 6-4...or 7-3, or even 8-2. If you did enough runs of ten flips, sooner or later you'd even get a run of 10 heads and no tails. Now if you flipped the coin 1,000 times.....there is virtually NO chance of you being more than a percentage point or two away from the 50/50 mark, because in *that* many trials the true odds would manifest themselves. In a hundred lifetimes you couldn't flip the coin a series of 1,000 times and get 650 heads or tails. It would simply *never* happen.But over only ten flips...you'll get 8-2 VERY often. Off the top of my head I calculate it would be one out of every 1004 series of 10 flips that you'd get a 10-0 result.
So in other words....if you got a thousand of your closest friends together, and all of you flipped a coin 10x, the odds say that ONE of you would get 10 heads....a handfull would get 9-1 and quite a number would get 8-2.
Now...a "trend" follower would walk up and down the line of 1,000 coin-flippers and find the guy who flipped 10 heads and no tails. "I've got a solid 10-0 trend here"...he'd say. "Every time Mr. Easymoney here flips the coin it's got a 100% chance of coming up heads........so on his next flip.....I want you to put 10x your normal bet on his flip coming up heads, because it's a 100%, 10-0 "Perfect Play"" (isn't that what Mark Lawrence calls his "trend" following plays?).
Now I hope there isn't anyone reading this who doesn't realize that Mr. Easymoney's chances of throwing a "head" on his 11th throw are EXACTLY 50/50. Out of 1,000 people flipping a coint 10x SOMEONE figured to go 10-0...and it just happened to be him....but that has *ZERO* predictive value for any future flips.
And by the same token, if you turn on your computer and let it plow through thousands of NFL games (or any other events, for that matter), looking for "trends".....*OF COURSE* it will find series of games that are outside the 50/50 envelope. You might find, for example, that teams from cities starting with "P" are 20-7 ATS when playing at home the week after losing on the road by between 10 and 14 points. Believe it or not, there are many "handicappers" out there who actually publish their picks with explanations no less idiotic than this example.
If you look at enough games, you'll find a MILLION "patterns" that are nothing more than the *inevitable* fluctuation that you get over the short run. Just as you probably won't get 5 heads and 5 tails if you take a coin out of your pocket right now and flip it 10 times.
If you look at all the teams that win by between 22 points and 27 points while playing in a dome who travel the next week........there's a very good chance that they will either cover more than 50%...or less than 50% of their next games.....just due to the random element that's always present. And if THOSE teams don't do somehing "odd", like go 40-20 ats the next week...you can be sure that if you look at ENOUGH situations you'll find the one that DID do something that SEEMS significant...maybe the teams that lost 2 weeks in a row by exacly 4 points turn out to be 23-3 ats the next week.
Just remember those 1,000 guys flipping coins......ONE of them *will*,in all probablility, go 10-0. But. He's no more likely than any of the other 999 guys to toss a head NEXT flip...and THAT is the issue that these "trend handicappers" can't seem to grasp.
So you get people who buy the trolling software, leave it on for days on end....and come up with a book full of "trends". And bingo...they are in the sports service business. The problem is, that what they have done is no different than walking up and down the line of 1,000 coin flippers and picking out the 30 guys who happened to go 9-1, and recruiting them to flip coins. "I've got 30 guys who are 9-1 here, all we have to do is bet on heads on every one of their flips and we can't go wrong". As loony as this sounds...it is EXACTLY the reasoning employed by most trend handicappers.
Now..........there *are* some "trends" that really ARE trends. For one....they MAKE SENSE. For example, it used to be you could make good money in the NBA just betting against teams that won at home and played on the road the VERY next day. The reasoning is obvious here...and it worked well for a while. The problem is that even LOGICAL trends only count as long as the linesmaker doesn't factor them into the line. Once Vegas catches on...the line is adjusted and the "trend" ceases to exist...because the factor that it identified is *already in the line*.
So the only way you will EVER make money from trends is first to discover one which ISN'T just random.....one that makes solid sense. Like the old 30-30 system that we all used to use. A team wins two weeks in a row and scores 30 or more in both games...bet against them the next week. The logic is pretty sound.....after those big wins they may have a let down. The public will overvalue them based on their two most recent games and push the line up a few points higher than it should be....giving you mathmatical value on the underdog. a sound system...based on a "trend".
*BUT*.....as soon as it becomes known to more than a few people...the linemaker will begin to adjust the line...and presto, the trend doesn't even EXIST any longer, since it was only EVER a trend relative to the pointspread...and if the pointspread is adjusted to ACCOUNT for the situation, the trend is voided.
Learn to differentiate between REAL patterns that teams fall prey to....and the millions of RANDOM patterns that crop up through the normal course of thousands of games being played.
Only by finding patterns that *make sense* can you HOPE to make money from trends. Teams that play on the West Coast on Monday...then fly East 3 time zones...for examle. If you found that they were, say, 40-50 ats the next week, you could make some sense of it and figure it was probably a pattern that may actually *repeat*.
Except...if the linemaker also spots it...the line will move a point or two and the "system" won't work any longer. So if you want to be a trend handicapper.....not only do you have to find trends that are statisically significant AND logical, but you have to find trends that the linemaker doesn't know about. And how likely do you think that is? Everyone has the same software spinning away day and night. THERE ARE FEW IF ANY *MEANINGFUL* PATTERNS NOT ALREADY IN THE LINE.
I found a silly little over/under "trend" many years ago that has been pretty good at predicting Unders in the NFL. I've never told one person how it works, though I happily give out the system's plays. Lots of them lose...but over several hundred plays it's ahead by a modest margin...and hits about 55%. Last week it went 2-0, and this week it has the under in the Houston/Jax game. But it could lose this, week, and for the next several weeks.....and everyone would jump ship, even though it's a long-term moneymaker. That's the thing, even with 55 white marbles in your jar and only 45 black, you can still pull 5 or 10 blacks in a row every now and then...at which point everyone thinks the "system" doesn't work any longer.
So the moral of the story is.....stop looking for silly little patterns that are almost ALWAYS the result of random fluctuation......they will lead to the demise of your bankroll. Fundemental handicapping, combined with SOME situational handicapping (e.g., team A had a BIG win last week against an old rival.....is playing Houston this week and is a 15 pt favorite, and plays it's other arch rival next week) is the only hope of overcoming the vig. But of course, a lot of these situtions are already in the line....because the linesmakers know about "let downs" and "sandwich" games too.
The value in the NFL is almost always in the DOG, because the public STAMPEDE onto the favorites....like Pitt last night. My stats showed Baltimore to be a 2pt favorite last night, though the two key inuries in the defense scared me off the moneyline. Now we all look at different stats, and come up with different conclusions even from the *same* stats...but what was everyone looking at last night? Almost all of the key stats were pretty much even. But the public couldn't bet enough on Pitt...and the line went so far North that today, in the cold light of dawn it looks like a joke.
Fundementals........let the suckers look for .....and make bets based on "trends".
There are 2 or three sports services that I actually PAY for, in addition to doing my own NFL handicapping. These guys are good, and do solid, fundemental and situational handicapping, and there is NO chance of one of them EVER saying "Wow, Atlanta are 5-0 when playing on the road the week after winning at home by less than 7 points". No real handiapper would ever say such a thing.
There are certainly much better handicappers in here than me. If last year was any indication, I could never come close to Franks NBA percentage...or Chuck's MBL. One guy I follow most closely in the NFL (and I pay for his service) is so good that if he puts out a play that goes against one of my own...even if I've already bet my own play I'll put a bet on his side and just lose the juice. The guy is simply better than I am at picking NFL games.
I've studied the game as much as I possibly can, and still struggle to do much more than 55%. So there are guys who have obviously discovered things I haven't...and thank god for that. I'm happy to follow other people's picks once I know they are solid.
Let's get one thing straight from the start....almost ALL of these "trends" aren't "trends" at all....they are minor anomalies....blips on the radar screen that mean NOTHING.
I'll see guys like Mark Lawrence or (god forbid) Brandon Lane (who I commented on at more length in the thread called "update on my friend") raving about how a perfect 6-0 trend applies to a given game. This demonstrates such a lack of sound mathmatical reasoning that it's fucking terrifying that people put their money at risk based on this kind of advice.
First, let's get to the bottom of what actually *causes* most of these *trends* that so many handicappers are so proud to have "discovered".
We all know that if you flip a coin 10x, the odds say that you will get 5 heads and 5 tails. There is NO question that those are the true odds.
BUT....BUT....that doesn't mean that on any given series of 10 flips you WILL get 5 of each. you may well get 6-4...or 7-3, or even 8-2. If you did enough runs of ten flips, sooner or later you'd even get a run of 10 heads and no tails. Now if you flipped the coin 1,000 times.....there is virtually NO chance of you being more than a percentage point or two away from the 50/50 mark, because in *that* many trials the true odds would manifest themselves. In a hundred lifetimes you couldn't flip the coin a series of 1,000 times and get 650 heads or tails. It would simply *never* happen.But over only ten flips...you'll get 8-2 VERY often. Off the top of my head I calculate it would be one out of every 1004 series of 10 flips that you'd get a 10-0 result.
So in other words....if you got a thousand of your closest friends together, and all of you flipped a coin 10x, the odds say that ONE of you would get 10 heads....a handfull would get 9-1 and quite a number would get 8-2.
Now...a "trend" follower would walk up and down the line of 1,000 coin-flippers and find the guy who flipped 10 heads and no tails. "I've got a solid 10-0 trend here"...he'd say. "Every time Mr. Easymoney here flips the coin it's got a 100% chance of coming up heads........so on his next flip.....I want you to put 10x your normal bet on his flip coming up heads, because it's a 100%, 10-0 "Perfect Play"" (isn't that what Mark Lawrence calls his "trend" following plays?).
Now I hope there isn't anyone reading this who doesn't realize that Mr. Easymoney's chances of throwing a "head" on his 11th throw are EXACTLY 50/50. Out of 1,000 people flipping a coint 10x SOMEONE figured to go 10-0...and it just happened to be him....but that has *ZERO* predictive value for any future flips.
And by the same token, if you turn on your computer and let it plow through thousands of NFL games (or any other events, for that matter), looking for "trends".....*OF COURSE* it will find series of games that are outside the 50/50 envelope. You might find, for example, that teams from cities starting with "P" are 20-7 ATS when playing at home the week after losing on the road by between 10 and 14 points. Believe it or not, there are many "handicappers" out there who actually publish their picks with explanations no less idiotic than this example.
If you look at enough games, you'll find a MILLION "patterns" that are nothing more than the *inevitable* fluctuation that you get over the short run. Just as you probably won't get 5 heads and 5 tails if you take a coin out of your pocket right now and flip it 10 times.
If you look at all the teams that win by between 22 points and 27 points while playing in a dome who travel the next week........there's a very good chance that they will either cover more than 50%...or less than 50% of their next games.....just due to the random element that's always present. And if THOSE teams don't do somehing "odd", like go 40-20 ats the next week...you can be sure that if you look at ENOUGH situations you'll find the one that DID do something that SEEMS significant...maybe the teams that lost 2 weeks in a row by exacly 4 points turn out to be 23-3 ats the next week.
Just remember those 1,000 guys flipping coins......ONE of them *will*,in all probablility, go 10-0. But. He's no more likely than any of the other 999 guys to toss a head NEXT flip...and THAT is the issue that these "trend handicappers" can't seem to grasp.
So you get people who buy the trolling software, leave it on for days on end....and come up with a book full of "trends". And bingo...they are in the sports service business. The problem is, that what they have done is no different than walking up and down the line of 1,000 coin flippers and picking out the 30 guys who happened to go 9-1, and recruiting them to flip coins. "I've got 30 guys who are 9-1 here, all we have to do is bet on heads on every one of their flips and we can't go wrong". As loony as this sounds...it is EXACTLY the reasoning employed by most trend handicappers.
Now..........there *are* some "trends" that really ARE trends. For one....they MAKE SENSE. For example, it used to be you could make good money in the NBA just betting against teams that won at home and played on the road the VERY next day. The reasoning is obvious here...and it worked well for a while. The problem is that even LOGICAL trends only count as long as the linesmaker doesn't factor them into the line. Once Vegas catches on...the line is adjusted and the "trend" ceases to exist...because the factor that it identified is *already in the line*.
So the only way you will EVER make money from trends is first to discover one which ISN'T just random.....one that makes solid sense. Like the old 30-30 system that we all used to use. A team wins two weeks in a row and scores 30 or more in both games...bet against them the next week. The logic is pretty sound.....after those big wins they may have a let down. The public will overvalue them based on their two most recent games and push the line up a few points higher than it should be....giving you mathmatical value on the underdog. a sound system...based on a "trend".
*BUT*.....as soon as it becomes known to more than a few people...the linemaker will begin to adjust the line...and presto, the trend doesn't even EXIST any longer, since it was only EVER a trend relative to the pointspread...and if the pointspread is adjusted to ACCOUNT for the situation, the trend is voided.
Learn to differentiate between REAL patterns that teams fall prey to....and the millions of RANDOM patterns that crop up through the normal course of thousands of games being played.
Only by finding patterns that *make sense* can you HOPE to make money from trends. Teams that play on the West Coast on Monday...then fly East 3 time zones...for examle. If you found that they were, say, 40-50 ats the next week, you could make some sense of it and figure it was probably a pattern that may actually *repeat*.
Except...if the linemaker also spots it...the line will move a point or two and the "system" won't work any longer. So if you want to be a trend handicapper.....not only do you have to find trends that are statisically significant AND logical, but you have to find trends that the linemaker doesn't know about. And how likely do you think that is? Everyone has the same software spinning away day and night. THERE ARE FEW IF ANY *MEANINGFUL* PATTERNS NOT ALREADY IN THE LINE.
I found a silly little over/under "trend" many years ago that has been pretty good at predicting Unders in the NFL. I've never told one person how it works, though I happily give out the system's plays. Lots of them lose...but over several hundred plays it's ahead by a modest margin...and hits about 55%. Last week it went 2-0, and this week it has the under in the Houston/Jax game. But it could lose this, week, and for the next several weeks.....and everyone would jump ship, even though it's a long-term moneymaker. That's the thing, even with 55 white marbles in your jar and only 45 black, you can still pull 5 or 10 blacks in a row every now and then...at which point everyone thinks the "system" doesn't work any longer.
So the moral of the story is.....stop looking for silly little patterns that are almost ALWAYS the result of random fluctuation......they will lead to the demise of your bankroll. Fundemental handicapping, combined with SOME situational handicapping (e.g., team A had a BIG win last week against an old rival.....is playing Houston this week and is a 15 pt favorite, and plays it's other arch rival next week) is the only hope of overcoming the vig. But of course, a lot of these situtions are already in the line....because the linesmakers know about "let downs" and "sandwich" games too.
The value in the NFL is almost always in the DOG, because the public STAMPEDE onto the favorites....like Pitt last night. My stats showed Baltimore to be a 2pt favorite last night, though the two key inuries in the defense scared me off the moneyline. Now we all look at different stats, and come up with different conclusions even from the *same* stats...but what was everyone looking at last night? Almost all of the key stats were pretty much even. But the public couldn't bet enough on Pitt...and the line went so far North that today, in the cold light of dawn it looks like a joke.
Fundementals........let the suckers look for .....and make bets based on "trends".
There are 2 or three sports services that I actually PAY for, in addition to doing my own NFL handicapping. These guys are good, and do solid, fundemental and situational handicapping, and there is NO chance of one of them EVER saying "Wow, Atlanta are 5-0 when playing on the road the week after winning at home by less than 7 points". No real handiapper would ever say such a thing.
There are certainly much better handicappers in here than me. If last year was any indication, I could never come close to Franks NBA percentage...or Chuck's MBL. One guy I follow most closely in the NFL (and I pay for his service) is so good that if he puts out a play that goes against one of my own...even if I've already bet my own play I'll put a bet on his side and just lose the juice. The guy is simply better than I am at picking NFL games.
I've studied the game as much as I possibly can, and still struggle to do much more than 55%. So there are guys who have obviously discovered things I haven't...and thank god for that. I'm happy to follow other people's picks once I know they are solid.
Comment