Math ?

[Padre]

Maybe i have been tossing numbers around all day but the life of me this makes no sense to me.

Ralph and Bob have two months of win/loss data.

Ralph
Nov. 39-35 for a winning % of 52.7%
Dec. 23-8 for a wining % of 74.19%

Bob
Nov. 26-22 for a winning % of 54.17%
Dec. 7-2 for a winning % of 77.78%


Now given this example Bob performed better in the month of Nov. 54.17%>52.7% and again in the month of Dec. 77.78%>74.19%.


Now if I were a betting man(which ,well, I am) and wanted to coattail someone then I would take Bobs picks right? they offer a better winning percentage each month(units excluded just % here)

But look if you combine both months(one would think that you would get a better % with Bob ,right)

Combined:

Ralph 62-43 win% 59.05%
Bob 33-24 win % 57.89%


What the hell happened? Did i brain fart and do my math wrong? just makes no sense to me.


HELP:director:

 

[smurphy]

Ralph has played a lot more games. Bob's 7-2 streak is nice, but it's only 9 games. His 26-22 prior absorbs the streak and brings the % down. Ralph's 23-8 run is much more impressive. ....My money is on Ralph!

 

[Sportsaholic]

Easy.................Bob likes beer :shrug:

 

[BobbyBlueChip]

Ralph has played a lot more games. Bob's 7-2 streak is nice, but it's only 9 games. His 26-22 prior absorbs the streak and brings the % down. Ralph's 23-8 run is much more impressive. ....My money is on Ralph!

Great job Smurph!!!

:toast:

 

[smurphy]

Great job Smurph!!!

:toast:

I'm 15-2 in December.

 

[Padre]

That is where i think i am getting confused.

u see i agree, but it comes down to the debate of,

do you want a higher win % over fewer games or take the lower win % over twice as many games.

u see if you were to use a money management system , similar to the Kelly Critrion model it is solely based on the win % , regardless of the amount of games.


like you Smurphy, i too like to think that 100-50 is great, but will that get ya 200-100 just the same. who knows.


http://www.albionresearch.com/kelly/default.php


u see i believe in the Kelly critrion model, i think that it is very good if you have a solid foundation in the projected win/loss percentages.

all the cons on such money management , have been because people have over estimated their %, and when to apply it.

when do i increase or decrease my bet size? is alot of when people do not relize how to prperply utilize it.

First , if u need to decrease bet size from ur original bet size, u are wasting time, and have overstated your win%.

second if your applying it after every wager , again wasting time, because the sample size is far too small to have an impact on the model.

what i have always done is short my win% by a few points, and wait until my bankroll doubles before i change unit size, and believe me, i have been able to maximize my profits, given this strategy. nothing else has worked for me.

 

[smurphy]

just go with ralph

 

[Padre]

Smurphy,

if Ralph goes on a losing streak , can i hold you solely responsible for my losses.:0corn

 

[PocketAces]

I believe in a flat betting system.

I would rather hit 54% on 10,000 decisions a year than hit 100% on one 5* or whatever.

Wouldnt you?

Don't handicap your handicapping by having 2* 3* 4* and 5* plays. If you figure you have a 52.5% chance of hitting it you should play it. Yor 2*s could become meaningless. Also you could be eliminating solid plays by being so selective.

Just my opinions. Not try to start anything here.

 

[Padre]

if interested, Ralph and Bob are just filters in my database.

I dont want people beating down some guy selling his services, named Ralph or Bob.

basically Ralph was totals on a game, and Bob was the side.

in short the logic is this, if a team has a double digit lead in an NBA game at the half, it does have a affect on the total for the second half. But just not have an impact on the side,

example: team A leads by 15, the dog nor the fav, had much impact on the outcome.

whereas, on the total it did. (i guess that would lead to the logical conculsion that the pace of the game is affected.) yet the leader didnt care much to how much they won by , just as long as they won.

 

[Padre]

I believe in a flat betting system.

I would rather hit 54% on 10,000 decisions a year than hit 100% on one 5* or whatever.

Wouldnt you?

Don't handicap your handicapping by having 2* 3* 4* and 5* plays. If you figure you have a 52.5% chance of hitting it you should play it. Yor 2*s could become meaningless. Also you could be eliminating solid plays by being so selective.

Just my opinions. Not try to start anything here.

I am with u 100% , i could not send this idea home any better.

i have always believed that if u have a system like this, why even play the lower stars. makes no sense.

PA, what i was saying ,evaluate Kelly to see what is your bet size. if your hitting 58% u can increase ur bet slightly thats all

not the ol , this game falls into my 52% fromula ,therefore i bet this.

no all games are valued the same. it is just at what point do u increase ur bet size if you are winning.

 

[PocketAces]

just go with ralph

I am with u 100% , i could not send this idea home any better.

i have always believed that if u have a system like this, why even play the lower stars. makes no sense.

PA, what i was saying ,evaluate Kelly to see what is your bet size. if your hitting 58% u can increase ur bet slightly thats all

not the ol , this game falls into my 52% fromula ,therefore i bet this.

no all games are valued the same. it is just at what point do u increase ur bet size if you are winning.

All good points. I'm really going to have to look closer at this.

Thank you for posting. Learned something.

 

[Padre]

PA, u see this is what i do,

no stars,no GOY, no bladder busters ,non of that non-sense.

i evaluate my win/loss % then minus 2 to 3 points. at the start of a season. in a certain sport.

Of course that win % is based on past seasons.

so for the NBA,for example, i would normally hit 57%.
i then put it into the formula. minus 2 points, so my bet size or unit size is 5.5%. now i sometimes think that this is still a bit exgressive , so i sometimes trim it back to just 5%.

Now this does not mean i take 5% of every bankroll balance daily, 5% of $1000 is $50. that is my bet size.

now i dont care about the best looking wager i have ever seen, i do not vary this bet size until,

i have doubled my bankroll, now sometimes this never happens in a course of a season or even a year. but in the event that it does, i then reevaluate my win % and bankroll size, therefor i should have doubled the $1000 to $2000 or more, i then recalculate and if it is still 5% then my unit size now increases to $100.

i hope this helps.

but this is what i found out to be the best way of handling the money management question, i just had problems earlier with a logical concept thats all.

 

[BleedDodgerBlue]

not many people know how to use kelly crit. correctly. you are basically wagering on a theoretical edge becuase you are giving yourself a percentage chance of that bet winning or percent "edge" you have. to get this percentage is still mostly a theoretical number.

as for growing your bankroll, with only 1k at risk, you should be more risk aversive. I know you have more than 1k and that if you go broke you aren't going to just sit out but rather reload. You should be coming up with your percentage to risk daily and uppping your plays accordingly. if you win around 3 units on day 1, you should be betting with a 1150 bankroll. alternatively if you lose three you should work of 850....etc...you aren't talking that much of a percentage increase and if you run hot for a while it will really show. kelly works to eliminate your risk of ruin if you play solely with this one bankroll although in theory if you really run bad, you could be betting $5 a game and it would take forever to make it worth your while.



gl

 

[Padre]

BDB , i too understand that,

but in using this daily or even on a weekly basis, u run into that Ralph Bob problem, ur sample size is too small, to predicate or calculate a win %, if you run hot for like 3 days and go 12-1, then ur bet size is going to be too high given the next run, which law of averages will bring u back down. so this is why i apply this model only after i have doubled my orginal bankroll.

for 2 reasons, i then have (hopefully) enough games to recalculate my win % or least to see how mush of an edge i might have.

Mind u, my unit size rarely changes much, as it should not, given my original win % and the amount of games involved.

what i was saying is Kelly works very good , or at least for me in maximizing my profits.

u see i have this one "system" (now i dont want to get into it, given the numbers, and no i will not explain it to anyone) but this thing hits about 86% of the time. but the pay out is 25%, thats 1 and 4 , not a good payout , but the winning percentage is what makes it worth it.

Now Kelly shows me given these numbers that i need to bet 30%, hey 3 losses and i am wiped out. But I trim it back to 25%. And i have used this for 8 years now, and I have spun the numbers and i seem to always come out with the best profits , i could have. The only time i have found this to "hurt" me is exactly the time that i double my backroll, then i hit a loss. Percentage wise, that loss just ate up a larger portion of my previous profits. But i do not recalculate every time, the "opportunities" just are not that many in one season, sometimes 1 or even 2 a week.

I just use it to determine about how much i should be wagering.

now i know u use your 2-5 star system, which works for you, i see your numbers, u do quite well.

but i just can not determine if one game is better than another. to me it is simply a play or not play.

if a game doesnt meet a certain criteria then i do not play it.

I know people that have 5 to 10 different stipulations on a game. if a certain team meets 4 of them, then it now becomes a four star play.

for me if it doesnt meet all of them , then why play it. for action? people that are in this, need to understand , are they in it for money or entertainment. because the "action" junkies are just going to shit it all away. and that is how i feel when it comes to a 2 star play vs. a 5 star play. to me the 2 star play is just there to get me to watch the game. why bother.:shrug:

i dont want to upset anyone, this is JMO, and how i handle MY money, how and what u do with yours is entirely up to u.

 

[BleedDodgerBlue]

but in using this daily or even on a weekly basis, u run into that Ralph Bob problem, ur sample size is too small, to predicate or calculate a win %,.

to each his own and whatever works for him or her......

who cares about the Ralph or Bob problem. In theory if you are established in a sport meaning over years you have beaten it, then it doesn't really matter what your name is... if you can hit 54 percent or thereabouts in any sport you want to make every possible play you can. its about volume, and not picking your spots. if you never put more than 20 percent of your alloted roll in play, you can play 3 or 4 percent plays and churn them out. streaks don't matter because at the end of the season you will have hit your 54 percent or thereabouts by the end of the season. that's why the nfl is such a different animal, there just isn't enough plays. I laugh at all the people who claim to make a living of the NFL....lol

for me, i use a 1-5 unit system (star makes it seems like you are selling something). 1 units for longshot moneylines, 4 and 5 rare, and usually between 2 and 3. it might or might not surprise you, but it takes me about 3 minutes max to cap an nba game and even that might be high. stats, trends, streaks, injuries mean nothing to me. the only stat of importance is the opening line. don't care how it moves really. i cap purely on feel and wager how strongly i feel about it. anyone who pours hours into stats and math, more power to them, but it won't help in the nba I can assure you that. if it works great, but you are wasting your time. I look at the opening line, and any injuries to star players (usually wagering against the team missing someone his first game out). that's it. same with baseball.....starting pitchers, opening lines and feel.

i guess in whole I agree with alot of your post, in that money management is the key within reason. some people are more risk averse than others. i tend to be more to the riskier side, but can afford to do so. as long as you have some kind of money management and can actually beat a game, you will have no problems. i've seen a lot of really good handicappers go broke. i've see even more great, great poker players on the rail as well. sports and poker take discipline.


gl

 

[smurphy]

to each his own and whatever works for him or her......

who cares about the Ralph or Bob problem. In theory if you are established in a sport meaning over years you have beaten it, then it doesn't really matter what your name is... if you can hit 54 percent or thereabouts in any sport you want to make every possible play you can. its about volume, and not picking your spots. if you never put more than 20 percent of your alloted roll in play, you can play 3 or 4 percent plays and churn them out. streaks don't matter because at the end of the season you will have hit your 54 percent or thereabouts by the end of the season. that's why the nfl is such a different animal, there just isn't enough plays. I laugh at all the people who claim to make a living of the NFL....lol

for me, i use a 1-5 unit system (star makes it seems like you are selling something). 1 units for longshot moneylines, 4 and 5 rare, and usually between 2 and 3. it might or might not surprise you, but it takes me about 3 minutes max to cap an nba game and even that might be high. stats, trends, streaks, injuries mean nothing to me. the only stat of importance is the opening line. don't care how it moves really. i cap purely on feel and wager how strongly i feel about it. anyone who pours hours into stats and math, more power to them, but it won't help in the nba I can assure you that. if it works great, but you are wasting your time. I look at the opening line, and any injuries to star players (usually wagering against the team missing someone his first game out). that's it. same with baseball.....starting pitchers, opening lines and feel.

i guess in whole I agree with alot of your post, in that money management is the key within reason. some people are more risk averse than others. i tend to be more to the riskier side, but can afford to do so. as long as you have some kind of money management and can actually beat a game, you will have no problems. i've seen a lot of really good handicappers go broke. i've see even more great, great poker players on the rail as well. sports and poker take discipline.


gl

<object width="425" height="344"><param name="movie" value="http://www.youtube.com/v/TtSdJeDpjkk&hl=en&fs=1"></param><param name="allowFullScreen" value="true"></param><param name="allowscriptaccess" value="always"></param><embed src="http://www.youtube.com/v/TtSdJeDpjkk&hl=en&fs=1" type="application/x-shockwave-flash" allowscriptaccess="always" allowfullscreen="true" width="425" height="344"></embed></object>