Cappin by trends and other nonsense

[dante1]

Before I piss anybody off and cause problems let me give a caveat, this is my opinion and I may be wrong. I am willing to listen to opposite views and if you use these methods and find success than I am wrong, maybe.

I think most if not all short term trends are completely uselss in cappin sports. I also think most long term trends are probably useless depending on the specific trend.

Let me give a hypothetical example. Okay, V Tech is 0-22 ATS when they play at night after a full moon. Tell me please what does this have to do with anything? But, some players will see this "trend," and use it as a cappin tool. When in reality the chances of V Tech covering in their next game is exactly even.

How about this trend? Again these are hypothetical situations, I am just making these numbers up. Okay, teams that are getting 2.5 points cover at 58.6% of the time and teams that are getting 3.0 points cover at 49.2%. Does, this mean that 2.5 dogs are better plays than 3.0 dogs. That is pure nonsense. It is a short term trend and it is not credible. If you could follow these numbers over say 10,000 games you will find that about 5,000 or close to that number end up on each side. Also, take into consideration the fact that some books will have this game at 2.5 and some will have it at 3.0.
So which spread is the one you use? A line variance is caused by the amount of action on a specific team. How can that fact cause or be related to a valid trend. Just doesn't make any sense.

My point is that if you critically examine short term trends I think you will conclude that they mean very little in cappin a game. I never even look at them.

Some might argue that the dog player, and I am a dog player is also a trend. And you might be correct, it may just be a short term trend. But this trend, I believe, has just a tiny bit of merit.

[pokernut9999]

Yea like an earlier post today said favorites only cover 49% in week 1. Guy said looks like a good week for the pups. Why if you bet all dogs or favorites you lose. Most trends mentioned are long term and have merit to them. When a trend has a 80 to 90% to it over 10 or more games it should not be ignored. Certain teams can not and will not beat certain teams.

[dante1]

Yea like an earlier post today said favorites only cover 49% in week 1. Guy said looks like a good week for the pups. Why if you bet all dogs or favorites you lose. Most trends mentioned are long term and have merit to them. When a trend has a 80 to 90% to it over 10 or more games it should not be ignored. Certain teams can not and will not beat certain teams.

Yes Michigan will never lose to Sisters of the Blind, but that isn't a trend. That is because Sistes of the Blind can't play football.

I don't think 10 or more games is enough to validate a trend.
If the Yanks beat the White Sox 10 times in a row after a loss you think that is a valid trend?

[Willie Bee]

Understand where you're coming from, Dante. I see a lot of 'trends' when editing and posting the front page plays from all the different cappers, and some of them don't make much sense to me. For example -- and I think this is something of a repeat from a thread conversation a week or so ago with EaglesPhan -- a team being 13-5 when they play on Wednesday. That doesn't rate as a solid 'trend' in my opinion. At the same time, I do recall a time when I lived in the Houston area and went to 40-50 Astros games a year, and the team was simply awful at home on Sunday. Was that because they had played the night before, was it because Chris Holt's turn in the rotation often came up on Sunday when the team was at home, or was it because they were mostly day games and the other team(s) just performed better during daylight hours (another trend)? I'm not sure, but it was what it was.

Look at it from the angle of a professional scout. I've met and talked to a couple of dozen MLB scouts in my life, those that were advance scouts and those that schlepped their way through tiny towns looking at high school kids. Some of them were what I'd call 'gut scouts,' meaning they'd have a little info like, "Quick first step, soft hands, uppercut swing, instinctive" written down in their little black books, and that would be it. Others would have a spiral notebook filled with the players history, how fast they got down to first from the batter's box, their 2-strike batting averages, range playing shortstop in night games on the road, whether or not they wore eye black or just sunglasses, all sorts of information. Different strokes and all that jazz.

[pokernut9999]

Yes Michigan will never lose to Sisters of the Blind, but that isn't a trend. That is because Sistes of the Blind can't play football.

I don't think 10 or more games is enough to validate a trend.
If the Yanks beat the White Sox 10 times in a row after a loss you think that is a valid trend?

Can't beat Ohio St either

Look at trends like Auburn vs Ga

Road team owns series for a long time
Last year GA reeling and Auburn #5 in nation

Ga crushes them on the road.

People study history-- history repeats itself

[dante1]

Understand where you're coming from, Dante. I see a lot of 'trends' when editing and posting the front page plays from all the different cappers, and some of them don't make much sense to me. For example -- and I think this is something of a repeat from a thread conversation a week or so ago with EaglesPhan -- a team being 13-5 when they play on Wednesday. That doesn't rate as a solid 'trend' in my opinion. At the same time, I do recall a time when I lived in the Houston area and went to 40-50 Astros games a year, and the team was simply awful at home on Sunday. Was that because they had played the night before, was it because Chris Holt's turn in the rotation often came up on Sunday when the team was at home, or was it because they were mostly day games and the other team(s) just performed better during daylight hours (another trend)? I'm not sure, but it was what it was.

Look at it from the angle of a professional scout. I've met and talked to a couple of dozen MLB scouts in my life, those that were advance scouts and those that schlepped their
way through tiny towns looking at high school kids. Some of them were what I'd call 'gut scouts,' meaning they'd have a little info like, "Quick first step, soft hands, uppercut swing, instinctive" written down in their little black books, and that would be it. Others would have a spiral notebook filled with the players history, how fast they got down to first from the batter's box, their 2-strike batting averages, range playing shortstop in night games on the road, whether or not they wore eye black or just sunglasses, all sorts of information. Different strokes and all that jazz.

You make an excellent point when you ask was it a valid trend or is there something else in play. If it was another variable then even though that trend seems to win it is still not a credible trend. Meaning you can't depend on it to be right more often than wrong.

Some trends may have a tiny bit of validity, I think most of them are total bs.

But, we can agree to disagree agreeably.

[dante1]

Can't beat Ohio St either

Look at trends like Auburn vs Ga

Road team owns series for a long time
Last year GA reeling and Auburn #5 in nation

Ga crushes them on the road.

People study history-- history repeats itself

Pokernut

I understand what you are saying. My argument is that if we could live long enough and watch the outcome of Auburn vs
Ga in thousands of games you will see many trends but none of them will be valid because at the end of a large number of games the W/L against the spread will be close to equal.

History repeats itself but probability has laws and those laws can't be changed by trends in sports.

[Willie Bee]

History repeats itself but probability has laws and those laws can't be changed by trends in sports.

Maybe in the long term, as you said. I mean, does it really matter right now what happened in the Auburn and Georgia contests back in the 1940s as far as how the two teams stack up this year? I would say no. But do trends have some value over the past 10 years? I think yes.

[dante1]

Maybe in the long term, as you said. I mean, does it really matter right now what happened in the Auburn and Georgia contests back in the 1940s as far as how the two teams stack up this year? I would say no. But do trends have some value over the past 10 years? I think yes.

Does this mean that if any team receiving 2.5 points in a game wins more often than a team receiving 3.0 points in a game this is a valid trend? This is a trend, and I think it is completely useless.

I will admit that some trends, if they deal with the actual skill involved in the game might be somewhat helpful. But, trends that have no relation to the actual playing of the game mean absolutely nothing.

[pokernut9999]

Pokernut

I understand what you are saying. My argument is that if we could live long enough and watch the outcome of Auburn vs
Ga in thousands of games you will see many trends but none of them will be valid because at the end of a large number of games the W/L against the spread will be close to equal.

History repeats itself but probability has laws and those laws can't be changed by trends in sports.

Only care about my lifetime not a 1000 years from now.

Hey Miss St covered once in 12 games vs LSU of course they were getting 33 and only lost by 31.

Lsu averages 42 vs them


So under 45 and St +18 looks good.trend is bogus im sure

[20Four7]

"Trends" are one tool in your arsenal.

There are several that come to mind for me.

1. The bears usually let down on a monday night game. No matter how good the bears seem to be they just don't seem to show up on Monday nights. That doesn't stop me from backing in the face of other information

2. The Jets always play Miami tough. I usually take a good hard look at the lines of these games since no matter who wins the game is usually close.

These are observations I've made and should probably go back and look at the stats of these.

I'm sure a lot of people look at certain trends such as result after a bye week, 1st home game, 1st away game etc. Again it's not a system where you "must" play but a took to be looked at in that game situation.

[dante1]

Only care about my lifetime not a 1000 years from now.

Hey Miss St covered once in 12 games vs LSU of course they were getting 33 and only lost by 31.

Lsu averages 42 vs them


So under 45 and St +18 looks good.trend is bogus im sure

Pokernut

This is just my opinion. If you have found success using these trends then obviously they work for you and that is great. I hope they keep working for you. I just personally don't buy into them.

Good luck whatever method you use. We just want to share info and help each other. Not cause hard feelings.

[Willie Bee]

Dante, I think you and I are on the same page with the issue. Some trends, especially when they're team specific or matchup specific, do seem valid to me. And other trends -- All teams favored by 3 win 58% of the time -- don't hold a lot of value with me. If the Yankees are 2-10 in Tampa the last two seasons, and both teams have essentially the same personnel, then that's a trend that I would think has some merit. But how the Yankees play in Tampa really holds no water with me as far as how the Pirates, Reds game will/should play out today.

[LT Profits]

I have posted several times that I sometimes use trends to either validate or eliminate a selection I have already made, but NEVER use them as the basis of a selection.

And even when I do look at trends, I eliminate the ones that are sheer coincidence, such as the previously mentioned days of the week. I only look at trends that have some sound logic behind them, such as a baseball team that is 14-40 vs. left-handed starters or a middle tier football team that has a great record vs. losing teams and a poor record vs. winning teams.

[dante1]

I have posted several times that I sometimes use trends to either validate or eliminate a selection I have already made, but NEVER use them as the basis of a selection.

And even when I do look at trends, I eliminate the ones that are sheer coincidence, such as the previously mentioned days of the week. I only look at trends that have some sound logic behind them, such as a baseball team that is 14-40 vs. left-handed starters or a middle tier football team that has a great record vs. losing teams and a poor record vs. winning teams.

LT

I can understand trends that have something to do with the play of the game. But, even with some of those you need to be careful. For instance, a trend says that Mich St has trouble beating Purdue at Purdue. Well if you think about it those kids turnover every four years. Now, each team has a different squad. You mean to tell me that this new squad of kids at Mich St will still have a propensity to lose at Purdue. Where is the logic? There is none, it is simply blindly following a trend.
Makes no sense.

[dante1]

And even if there is the slightest bit of validity in one team not being able to cover against another team how long do you think it will take the line maker to compensate? Not very long. Also we are not talking about a team beating another team we are talking about a team covering the line.

[pokernut9999]

95% of the lines come from power ratings with the adjustment of like 3 pts for home team. Has been that way for years. Never seen it any different.

[dante1]

95% of the lines come from power ratings with the adjustment of like 3 pts for home team. Has been that way for years. Never seen it any different.

Yes, and those power lines are adjusted constantly by
strength of team and how that team plays, not whether or
not they lose on the 4th Sunday when playing on the road.

[maritime]

I think, as a handicapper, you need at least to be aware of trends. I certainly agree that most trends are too short term to be "statistically significant."

What gets my attention is when a well known trend seems to conflict with the line set by oddsmakers. Here is a good example off the top of my head, though from last baseball season. We all know how Roy Oswalt dominates the Reds. Currently he is 19-1 lifetime against them. Generally this is reflected in the lines when Oswalt faces them, and there is a lot of chalk. However, when he faced them last season one time, the line, to me, was unusually lower than expected and even moved toward the Reds as gametime approached, for no other obvious reason. So I played the Reds, and won.

Generally, if a trend is significant, it will accurately be represented by the line, or if anything, overinflated by the line, so I don't think basing your bets on trends is worthwhile. Except in cases similar to above, where the line actually conflicts with the well-known trend.

[Willie Bee]

I think I know exactly what game you're talking about, maritime, early in the season and on the road in Cincy, and it wasn't like he was facing Arroyo or Harang, can't remember who started for the Reds right now. Houston almost came back and won it in the 9th as I recall against Stormy Weathers.

[dante1]

I think, as a handicapper, you need at least to be aware of trends. I certainly agree that most trends are too short term to be "statistically significant."

What gets my attention is when a well known trend seems to conflict with the line set by oddsmakers. Here is a good example off the top of my head, though from last baseball season. We all know how Roy Oswalt dominates the Reds. Currently he is 19-1 lifetime against them. Generally this is reflected in the lines when Oswalt faces them, and there is a lot of chalk. However, when he faced them last season one time, the line, to me, was unusually lower than expected and even moved toward the Reds as gametime approached, for no other obvious reason. So I played the Reds, and won.

Generally, if a trend is significant, it will accurately be represented by the line, or if anything, overinflated by the line, so I don't think basing your bets on trends is worthwhile. Except in cases similar to above, where the line actually conflicts with the well-known trend.

Can't argue with this. My premise was that most trends are
worthless. However, if a player uses them he must also use
some common sense and ask himself if the statement has anything to do with how the teams will play. Many times they are totally unrelated. There is no gambling god that makes a particular team lose on a particular day of the week or any such nonsense like that. Yet, some players use these as real tools. That was my only point.

[pokernut9999]

I dont buy in the day of week and bye weeks and crap like that. I do go by team vs team trends.

[durito]

There are all sorts of league wide long term trends that have been very predictive in football and basketball, college and pro.

But, I'm talking stuff that goes back 20 years.

[dante1]

Teams change constantly, thus talent changes constantly.
If talent changes constantly short term trends in evaluating who will cover is not relevant. But, if you mean that team A has all their players back and they killed team B last year who also has all their players back so team A is the better play, well I agree. But, not because of a trend saying they beat this team 3 years in a row. No, but because they still have the same team that beat this team 3 years in a row. That isn't a trend, that is superior talent.



IMO, and only talking about winning the game outright the better team usually wins the game. Now, not always sometimes luck comes into play and a team with little talent wins a game they shouldn't have. Remember NCState/Houston. But talent decides games, the linemaker attempts to even plays by convincing 50% of the players that the line will give the dog the advantage and 50% of the players that the line favors the chalk.



So talent wins games, the book tries to even that talent by making a line. Superficial trends are erratic and not valid statistically.

[durito]

I'm not talking about team trends, rather league wide trends. Ideas such as playing against teams on a spread losing streak, or playing on home dogs after an upset win (this is 57% since 1985 in the NFL).

There are tons of trends like this out there.

[dante1]

You know who I think can address this question better than all of us, a mathematician with a strong background in stats. I wouldn't be surprised if we have one or two on this forum.
If so, I would love to hear your response to this and I certainly
won't argue with you

[maritime]

his name is ganchrow, math whiz

[dante1]

his name is ganchrow, math whiz

Maritime, thank you he would be the guy. He is probably reading these posts and either LHAO or just totally disgusted with our ignorance. Maybe, that is why he didn't respond.
Those math guys could get cranky. (lol

[statguy]

JMO, but trend = expected, random fluctuation around the mean. The only thing 'stranger' than flipping a coin for a while and getting many heads or tails in a row, would be to flip a coin for a long time and get NO streaks at all. You guys probly slept through the class, but that bell curve is an amazing thing.

[dante1]

JMO, but trend = expected, random fluctuation around the mean. The only thing 'stranger' than flipping a coin for a while and getting many heads or tails in a row, would be to flip a coin for a long time and get NO streaks at all. You guys probly slept through the class, but that bell curve is an amazing thing.

I thought I was kinda saying this.

[statguy]

I know. I just said it better.

[dante1]

I know. I just said it better.

MUCH

[pokernut9999]

Last 6 years avg score LSU 42-7

Oh well another trend continues

[dante1]

Last 6 years avg score LSU 42-7

Oh well another trend continues

When a good team plays a shit team you get those
scores year after year. That isn't a trend that is better
talent. You confuse the two.

Tell you what my friend, I challenge you and not after the
games but before them. You post 6 or 8 games that you
will play because of trends. We will follow this for the
year and observe how your trends do. Doesn't have to be
6 or 8 any number you like. Explain your trend make your
selections and at the end of the year we will talk.

[pokernut9999]

covered 12 of last 13 years vs spread. What the heck do you want.