The Big Card

[therber2]

Vinny did you find any trends for favorites win 5 or more home/away games? or was this analyzed and found not to be worth it. Another poster here said that he was going to consider how to bet streaks with overs and unders. I'll look into how he's doing.

[therber2]

like the thread.
keep it up!

Thanks SA.

[therber2]

The only thought I have is to make the spreadsheet vertical instead of horizontal for each team.

Take a look at the streaks on the Totals too. Maybe we can glean something out of them too.

~ VV

A note to all who are doing lots of back testing. Just wanted to let you all know about for not letting you know sooner.

[therber2]

So for those who are looking to study an over or under streak you could query something like p:O and pp:O and ppp:O and pppp:O and ppppp:O and pppppp:O which would return games where the team's game total went over 6 times in a row.

[sycoogtit]

Note to self - How did Sycoog interpret my explanation?

Hey veni, I'm trying to backtest your system for 07-09. I've written the program and am comparing the 2010 results with yours to make sure it's running correctly. I'm not getting the same results, so I want to make sure I've got the rules right.

1) Each team must have played 8 games previous to their matchup
2) Whichever team has more wins over their last 8 games qualifies as 1 of 3 categories:
Home team with odds >= -103 is HOME DOG, PLAY +1.5
Home team with odds <= -105 is HOME FAV*
HOME FAV* with odds <= -150 is NO PLAY
HOME FAV* with odds > -150 is HOME FAV, PLAY ML
Away team with odds <= -105 is NO PLAY
Away team with odds >= -103 is AWAY DOG, PLAY +1.5

Is that right?

syco

8 previous games away or at home

rest looks good to me

Thank you sir.

What else have you missed by not reading the posts?

[sycoogtit]

Vinny did you find any trends for favorites win 5 or more home/away games? or was this analyzed and found not to be worth it. Another poster here said that he was going to consider how to bet streaks with overs and unders. I'll look into how he's doing.

Hey therb, I also tried this approach and it was worse than just requiring them to win more than the other team. I tried requiring the team in question to win 50% or 70% (two different runs of the program) of their last WINDOW_SIZE, and I also tried doing that and ignoring the win-loss of the other team.

I tried a kagillion variations, but the best results were the original version, which I guess is being called the morphed version.

Since the AD were performing so poorly in both RL and ML in all of my variations, I tried one more version yesterday. Make the window size 0 and ignore previous win-loss. This was really just to see how Away Dogs perform in general over the last couple years. For 08-09 combined, AD RL were -150 units and AD ML were -60 units.

[Vinny Vidivicci]

Saturday, May 22 -

WINDOW 8 Leans (in order of starts) -

Angels(4) @ St. Louis(4) - N/P
Florida(3) @ White Sox(3) - N/P
Balt.(3) @ Wash.(4) - Washington(1)
S.Fran.(2) @ Oak.(5) - Oakland(3)
Colorado(2) @ K.C.(4) - Kansas City(2)
Milw.(4) @ Minn.(5) - Minnesota(1)
Atlanta(5) @ Pitt.(2) - Atlanta(3)
T.B.(4) @ Houston(4) - N/P
Det.(3) @ Dodgers(6) - Dodgers(3)
Yankees(4) @ Mets(4) - N/P
Cincy(4) @ Cleve.(1) - Cincinnati(3)
Cubs(2) @ Texas(7) - Texas(5)
Boston(2) @ Philly(5) - Philadelphia(3)
Toronto(3) @ Arizona(3) - N/P
San Diego(5) @ Seattle(3) - San Diego(2)

~ VV

[Vinny Vidivicci]

Sycoog -

I have and I do read all the posts. I meant no offense when I wise-cracked to you a few days ago, and I mean no offense to you in what I'm about to say now.

When you said -

Hey veni, I'm trying to backtest your system for 07-09. I've written the program and am comparing the 2010 results with yours to make sure it's running correctly. I'm not getting the same results, so I want to make sure I've got the rules right.

1) Each team must have played 8 games previous to their matchup
2) Whichever team has more wins over their last 8 games qualifies as 1 of 3 categories:
Home team with odds >= -103 is HOME DOG, PLAY +1.5
Home team with odds <= -105 is HOME FAV*
HOME FAV* with odds <= -150 is NO PLAY
HOME FAV* with odds > -150 is HOME FAV, PLAY ML
Away team with odds <= -105 is NO PLAY
Away team with odds >= -103 is AWAY DOG, PLAY +1.5

Is that right?

I really didn't understand your breakdown above, so when Peterpan said -

syco

8 previous games away or at home

rest looks good to me

I assumed (I know, it's a bad word) that we were all on the same page in understanding the thesis behind this so-called system or angle. Yet, late last night Thurber was thinking it was simply the last 8 games, without regard to Home or Away. This gave me cause for concern because I thought everyone really understood it. If you were already in sync with the concept of it, GREAT. Really, I meant no offense a few days ago when I said "I don't know - they're YOUR graphs", and I meant no offense when I wondered how you had interpreted the system/angle. For what it's worth, the picks that Thurber posted (Post#699) differ from the ones I posted above(Post#707).

Respectfully submitted,

~ Vinny Vidivicci

[Vinny Vidivicci]

Sycoog-

To answer your question - "What else have you missed by not reading the posts?"

I've probably missed a lot - but not because I haven't read all of the posts. It's because my programming skills are limited to what I can type in by hand on a spreadsheet. You guys have number-crunching skills that I don't have. I took EVERY Math course in High School and College (circa 1963-1971, for those that are interested) except ONE - Statistics. And I really wish I had - given my interest in the analysis of sporting events. They didn't have computer programming back in the Dark Ages when I went to school. I bought an Atari 400 in the early '80s and taught myself BASIC. I wrote a program (in BASIC) that I took from a book to predict NFL outcomes. Not to digress, but the system was entirely stats-driven, where you took each team's stats and adjusted them to their level of competition in a ratio to the league average. You then adjusted THOSE numbers a 2nd time to account for the opponent's level of competition ratioed to the league average. The guy who wrote the book explained that using a calculator and a pad of paper it would take 6-7 hours to do all of the games each week. I did it the whole year and it picked 60%. Once the season was over, I bought my first computer. Updating and inputting the data took about 2 hours and 5 minutes later I had my picks. I saved 5 hours and millions of skin cells on my fingers!! After the first 3 weeks the NEXT year the system was about 40%; I started doing it by hand to see if my program had a bug, which now took me 8-9 hours between the computer and the calculator, and there was no bug. Anyhow - I hope you all enjoyed this story. In summary - that's the extent of my programming abilities.

~ VV

[cgars1]

keep workin it looks good

[Vinny Vidivicci]

8 Game Window Plays (MatchBook Lines) -

Washington -137 1 Unit
Oakland -107 1 Unit
Kansas City +106 .33 Unit
Kansas City +1.5 Runs -148 .67 Unit
Minnesota -131 1 Unit
Cincinnati +1.5 Runs -1.91 .67 Unit
Texas -142 1 Unit
Dodgers -133 1 Unit
Philly -116 1 Unit

Kansas City is a play against an AF losing 5 or more in last 8 - Record is 12-8 ML and 15-5 +1.5 RL, so I split the bet taking .33 ML and .67 +1.5 RL.
Cincinnati is a play against a HF losing 5 or more of 8 - Record is 17-10 +1.5 RL but the juice is high so I backed it down to a .67 Unit bet.

Good Luck -

~Vinny Vidivicci

[Devon]

Q about the Pythageron expectation.....

Do you use the numbers on current runs score and given up at this point in time... like right now? Basically, if a team scores much more than they give up they're have a higher expectancy to win?

Also, once you have numbers up... how do you decide which team would be the play?

[therber2]

Lol. To Vinny and Sycoog: Let's not let some miscommunications happen here; I know that neither of you guys meant disrespect to each other right?

I'm sorry Vinny; I am fairly certain sycoog understands (most likely better than I do) what you are doing. I should apologize; you guys are both working so hard, and I got preoccupied with some stuff, and asked a thick headed question because I didn't read everything that you posted.

So anyways, Vinny; most of my initial leans for today were favorites, and I was trying to find some matches with your system and mine to sort of find 1 really good pick. I love a dog system I have to say, but it might eliminate some good play on fav. situations. I'll play around with some queries to see if I find good fav. situations based on last 8 games.

The other thing that I am wondering (and if you already explained this I sincerely apologize :-) ) is how you react to the last 8 game record of the opposite team of your pick. For fades obviously you know what the opposite's record is, but then what is the side you fade to is only 1 out of 8?

[Vinny Vidivicci]

Devon, let me give you an example of applying the Pythagorean theorem using Runs-scored and allowed - and I'll use the Yankees and the Mets as an example:

Yankees scored 239 allowed 170
Mets scored 186 allowed 180

Yankees Runs Scored squared ................................. 239 times 239 = 57121
Yankees Runs Allowed squared ............................... 170 times 170 = 28900
Total these ........................................ ........................................ = 86021
Yankees Expected Win % is 57121 divided by 86021 ....................... = .664

Mets Runs Scored squared ....................................... 186 times 186 = 34596
Mets Runs Allowed squared ..................................... 180 times 180 = 32400
Total these ........................................ ........................................ = 66996
Mets Expected Win % is 34596 divided by 66996 ............................. = .516

Once you've done your calculations, you take the team with the HIGHER Expected Win % and divide it by the LOWER Expected Win % to make your own Odds on the game.

Yankees Expected Win % ............................... .664
Mets Expected Win % .................................... .516
Divide .664 by .516 ....................................... 1.287 Favoring the Yankees

YOUR Money Line on the game would be Yankees -129 on this game. Depending on how large of a difference you need to make a bet, you might take the Yankees at -110 or take the Mets at +150.

~ Vinny Vidivicci

[therber2]

Q about the Pythageron expectation.....

Do you use the numbers on current runs score and given up at this point in time... like right now? Basically, if a team scores much more than they give up they're have a higher expectancy to win?

Also, once you have numbers up... how do you decide which team would be the play?

Hey Devon, first off I run lots and lots of calculations, and then after long arduous hours of work I go with my gut I'm very undisciplined.

I used to use the pythag expectation for betting on College basketball going off the Kenpom site, but pretty new to doing it with MLB. Member Solobass just reminded me of it in a very inspiring discussion that he conducted at a place that I for various reasons that I honestly don't care much about, cannot mention here.

I quickly realized how I could apply some things I look at all the time into the equation. I know how most cappers think about the pythag exp. What I am trying out is an alteration of the formula that I think would be more accurate/telling (yeah I know...who am I to challenge the long lasting formula tried and true). It is an experiment; what can I say. I lost 50 units the other day; most of what I spew is probably non-sense.

"Basically, if a team scores much more than they give up they're have a higher expectancy to win?"

A. Depends. The theory is that if a team is scoring more runs than they are letting go that is good. As for having a "higher" exp. to win%, like I said that depends, and it is theoretical. If a team's record is .500 and their exp. % is .575 IN THEORY that means that they should be winning a bit more than they have. Take it with a grain of salt though. If that same team has a very low RPI then it would mean to me that they are a sloppy team; if that same team has a very high RPI then I would say that they are better than their record indicates...

[Vinny Vidivicci]

Therber -

Glad you're awake; I saw you burned the night oil last night.

I'm cool with everyone here, nor do I mean to disrespect anyone here.

When Sycoog first asked me about the parameters involved and he used > and < , etc. in his question, I had difficulty understanding it all and when Peterpan jumped in and answered for me I let it go as being answered (since I wasn't really understanding the question myself).

I'm just about out the door to go to work again (No, I'm not Peter Gammons, but I wish I was), but I posted something on one of the previous pages where I designated "Categories" for each part of the system. I'm working on stream-lining this for clarity and record-keeping, because I mentioned there were a few "Categories" that were lucrative. I'll post this here, or if anyone prefers, I'll e-mail them; so as not to clutter this thread. I'm sorry I made all of those in-game posts about Toronto last night; I wish I could delete them because they just clutter the thread.

2 Points I'd like to make -

POINT 1 -You asked about whether we had a game that we both agree on to find 1 really good pick. I don't agree with the 1 really good pick theory. If you want grass to grow you throw out a lot of grass seed. I'd rather take my chances on 5 to 9 bets then I would on 1 on any given day. I'm sure a lot of people loved Tampa Bay the day they got no-hit for instance. That can happen when you isolate your selections to 1 really good pick. If your handicapping has a POSITIVE expectation you want as many games as possible going - kind of like compound interest.

POINT 2 - I really think their is value in taking a MEDIAN of all 30 team's year-to-date 8 game window totals HOME and AWAY. This will give you a Power Rating on a 0 to 8 scale for all 30 teams, Home and Away. When a team has a substantial variance below their Power Rating, they are UNDER-achieving (which may make them DUE to win ... or not, we don't know yet). Likewise, OVER-achieving teams might be worth fading. I really feel in my gut this is something to look into.

~ Vinny Vidivicci

[therber2]

Sycoog-

To answer your question - "What else have you missed by not reading the posts?"

I've probably missed a lot - but not because I haven't read all of the posts. It's because my programming skills are limited to what I can type in by hand on a spreadsheet. You guys have number-crunching skills that I don't have. I took EVERY Math course in High School and College (circa 1963-1971, for those that are interested) except ONE - Statistics. And I really wish I had - given my interest in the analysis of sporting events. They didn't have computer programming back in the Dark Ages when I went to school. I bought an Atari 400 in the early '80s and taught myself BASIC. I wrote a program (in BASIC) that I took from a book to predict NFL outcomes. Not to digress, but the system was entirely stats-driven, where you took each team's stats and adjusted them to their level of competition in a ratio to the league average. You then adjusted THOSE numbers a 2nd time to account for the opponent's level of competition ratioed to the league average. The guy who wrote the book explained that using a calculator and a pad of paper it would take 6-7 hours to do all of the games each week. I did it the whole year and it picked 60%. Once the season was over, I bought my first computer. Updating and inputting the data took about 2 hours and 5 minutes later I had my picks. I saved 5 hours and millions of skin cells on my fingers!! After the first 3 weeks the NEXT year the system was about 40%; I started doing it by hand to see if my program had a bug, which now took me 8-9 hours between the computer and the calculator, and there was no bug. Anyhow - I hope you all enjoyed this story. In summary - that's the extent of my programming abilities.

~ VV

Thank Vinny Did enjoy the story. You're a bit older than sycoog and I we really appreciate what you've brought here. You've taught me quite a bit sir; it is much appreciated.

Actually once I automate things I sort of lose track of what it is I am really looking at. There is something about writing it out by hand the old fashioned way. I try to make a point to do at least one game a day on a piece of paper just so I know that I am not deviating from what works for me.

I guess that you could say that I am in a similar boat as you as far as the "number crunching." I am an Industrial Designer/Engineer; I work with manufacturing and design. I have a ton of respect for sycoog for what he is able to do; I know I wouldn't be able to do that. Funny thing Vinny, when I was very young (around 14 or something) I made a stock program in QBASIC which would tell you whether or not to buy a stock based on a range of PE ratios. That is really as far as I ever wanted to take it. In the last month though I have learned boatloads with excel. The program never ceases to amaze me.

[therber2]

Vinny I think my gut is telling me the same thing on point two there Just wondering how to do this efficiently; yes I was I burning the night oil and I am wired on coffee right now. One thing: calculating how many wins due or losses should be due will only result in you chasing down the same bet. Your first point was that I shouldn't try to find 1 really good bet. Not sure about that. I look for situations where the team I like is both better, and due for a win; while the opposing team should be drying up off a hot streak. On top of this I go off of me and sycoog's projected winner formula, and a whole slew of stats such as recent bullpen, and hitter stats. Now if one of your situations checked with this I'd say it was just another great factor for the pick, and then maybe add a unit to what I've got. I'll try to point out when I have a pick where the stars align. None today; however, I think that there may be room to make a few catagories where you are actually taking a favorite (other than the less than was it -140 fav?), as you say further stream-line would be great.

Side note: I don't give hoot about "clutter" guys. In game chit chat is all good. Post as you please; just keep in mind the OP So far I haven't had a single problem with anyone or anything in the thread.

[therber2]

Hey therb, I also tried this approach and it was worse than just requiring them to win more than the other team. I tried requiring the team in question to win 50% or 70% (two different runs of the program) of their last WINDOW_SIZE, and I also tried doing that and ignoring the win-loss of the other team.

I tried a kagillion variations, but the best results were the original version, which I guess is being called the morphed version.

Since the AD were performing so poorly in both RL and ML in all of my variations, I tried one more version yesterday. Make the window size 0 and ignore previous win-loss. This was really just to see how Away Dogs perform in general over the last couple years. For 08-09 combined, AD RL were -150 units and AD ML were -60 units.

Thanks for running it sycoog. Appreciate the hard work brutha. What I am thinking is that the favorite might be the favorite just because he is on the win streak, and logically worth fading.

How many games came up where there was a favorite with 4 or more wins out of the last 8 in say 2009 and/or 2008?

Here is my hypothesis for what might be a good "catagory" or morph (why are they called morphs?)

Catagory/Situation X: play on favorite (home or away not sure) with 4 or more wins out of the last 8 games. team must have a record% .500 or greater.

I don't want to put on more filters than the .500 or greater, but if that doesn't bring up good results then additional filters may be: .5 or greater plus higher RPI than opposing team. Team's exp. win% must be higher than their actual or opponent. Opposing team must have a less than .500 win%

Going...to query this if I can figure out that damn language. heh

[sycoogtit]

Okay enough drama in this thread. Sorry I got pissy, I just felt like I was the only person who gave a crap about the work I was doing to try to find something favorable from this system. I'm glad there's been more discussion about it and less of me posting graphs.

What I am thinking is that the favorite might be the favorite just because he is on the win streak, and logically worth fading.

Weird that you say that therber. I was just thinking along the same lines, but more generally. Whenever I try to find trends, I'm trying to find teams that SHOULD win the game and what would happen if I bet on them. What if there's value in finding teams that SHOULD LOSE the game (based on whatever logic you deem appropriate), but the books say they're actually the favorite?

Here is my hypothesis for what might be a good "catagory" or morph (why are they called morphs?) Catagory/Situation X: play on favorite (home or away not sure) with 4 or more wins out of the last 8 games. team must have a record% .500 or greater.

The 50% win rate is a variation I tried without success -- also tried 70%.

[Vinny Vidivicci]

I honestly appreciate what you've been doing for me(us) with the programming and graphs Sycoogit.

I mentioned to Thurber in the post I made right before I went to work on Friday that I was playing with some numbers in his winning expectation spreadsheet based on run-scoring innings and runs-allowed innings. It's early, but I found some "flukey", against-the-trend results since last Wednesday. Using this reverse-logic idea it was -

4-1 Wednesday
3-0 Thursday
4-0 Friday
2-1 Today

It's based on creating your own Money-line using the Exp. Win % column (Column H) in Thurber's Innings Pythag (the one you sent me Thurber). I wrote a quick formula to convert each team's numbers from Column H and adjust for Home Field advantage and create my own Money line on each series. I use this Money line against the real Money lines (at our books) and when an overlay of .33 or more occurs the results are the reverse of the records above. Yes, it picks LOSERS.

Now that you stopped laughing, could you pick 11 of 13 losers, if you tried to?

And, it kind of makes sense to me in a controlled-chaotic way. Vegas WANTS to put out lines that tempt wrong-way action. The old adage "If it looks to good to be true ...it probably is."

I'll put these picks out tomorrow also.

My Window 8 picks went 5-3 today, I made about 1/2 unit profit.

~ Vinny Vidivicci

[therber2]

twins -128 (3 units)
mil/min o9 (.5 units)
kc +1.5 -150 (1 unit)
col/kc u9.5 (1 units)
[b]fla +1.5 -177 (.5 units[/b])
fla +116 (.5 units)
cws/fla u8.5 (1.5 unit)
cin +1.5 -150 (1 unit)[/b]
cin/cle u8.5 (.5 unit)
dodgers -145 (1.5 unit)
det/los o8 (2 units)
sfo -110 (1 units)
sfo/oak o7 (3 units)
[b]atl -141 (3 units)
atl/pit u9 (.5 units)
was -140 (.5 unit)
padres -108 (3 units)
cub/tex u9.5 (.5 unit)

12-6 +11.9
Back on track boys

[therber2]

Okay enough drama in this thread. Sorry I got pissy, I just felt like I was the only person who gave a crap about the work I was doing to try to find something favorable from this system. I'm glad there's been more discussion about it and less of me posting graphs.



Weird that you say that therber. I was just thinking along the same lines, but more generally. Whenever I try to find trends, I'm trying to find teams that SHOULD win the game and what would happen if I bet on them. What if there's value in finding teams that SHOULD LOSE the game (based on whatever logic you deem appropriate), but the books say they're actually the favorite?



The 50% win rate is a variation I tried without success -- also tried 70%.

Good we are sort of on the same page with similar thought process. My question before I look into more detail on a favorite autobet series is: how many plays came up with the 50% (of 8 games) circumstance. If it was a fairly large amount I am going to consider filters.

[therber2]

keep workin it looks good

Hey cgars1. Thanks for poppin in.

[therber2]

Big card

early leans progress

TEX -170 (1.5 units)
CUB/TEX u9.5 (1.5 units)
STL -130 (.5 units)
DET/LOS o8 (.5 units)
MIN -200 (1 unit)
SEA -138 (1 unit)
SFO/OAK o7.5 (1 unit)
FLA -140 (3 units) *cws v. cle
CWS/FLA over8.5 (.5 units)
WAS -120 (.5 units)
NYY -132 (2 units)
NYY/NYM u7 (.5 units)
KC -140 (.5 units)
TAM -180 (3 units)
TOR -115 (2 units) *laa
TOR/ARI u9.5 (.5 units)
PIT +120 (1 unit)
ATL/PIT u8.5 (1.5 units)
CLE +126 (1 unit)
CIN/CLE o9.5 (.5 unit)

[Vinny Vidivicci]

Nice day, Therber.

[therber2]

Nice day, Therber.

Thanks Vinny, what do you guys have for today?

[Vinny Vidivicci]

I'll have the Window-8 picks done in a little while.

The reverse-logic picks from your projected winning % spreadsheet (11-2 since wednesday) -

Baltimore
Philly
Texas
Florida
K. City
Minnesota
Seattle
Yankees

I'll be playing all of the above, either ML or RL, or a combination of both.

~ VV

[therber2]

I'll have the Window-8 picks done in a little while.

The reverse-logic picks from your projected winning % spreadsheet (11-2 since wednesday) -

Baltimore
Philly
Texas
Florida
K. City
Minnesota
Seattle
Yankees

I'll be playing all of the above, either ML or RL, or a combination of both.

~ VV

Nice. GL today Vinny. Could you clarify the bolded? What exactly are you refering to when you say "reverse logic?" Do you mean fading the line dictated primarily by the streak?

[Vinny Vidivicci]

Thurber -

Post #721 explains it.

No, it's completely apart from the last 8 game system we have been working on.
I mentioned in a quick post as I was going out the door on Friday that I was playing with some numbers with your expectation win % spreadsheet based on run-scoring inning %.

~ VV

[Vinny Vidivicci]

Sunday Window 8 Picks -


Washington
Houston
Oakland
Arizona +1.5 RL
Mets +1.5 RL

Sometime today I am going to post a simplified explanation and an easier way to compartmentalize and evaluate the picks on the Window-8 System. I will include year-to-date records that will simplify which parts of the system are of value.

The idea that I'm really excited about right now gave me the picks that I listed in post #728 which has a record of 11-2 since last Wednesday. I'll put a more detailed explanation of this on my next post after I've gotten my bets in at MatchBook.

~ Vinny Vidivicci

[therber2]

Aaahhh. Got it. I actually skimmed/missed this post, but just read it now, and I am liking it quite a bit. That win rate is so high it seems too good to be true so I question a fluke; however, I see the logic.

Could you send me your sheet with the conversion formula with home adv. that you whipped up? Let me get it straight: you are using the STD exp win% (pythaginn version). Then you convert that win percent to a ML lets say -150. If the vegas (average, open, or close?) line is -117 that pick is the "loser?"

This is great Vinny. Shoot the moon with losers. I love it.

[Vinny Vidivicci]

Thurber -

The reverse-expectation picks from your projected winning % spreadsheet (11-2 since wednesday) -

Explanation -

In your Offensive/Runs-Scored and Defensive/Runs-Scored spreadsheet on Column H you computed each team's expected Win %. I used the numbers from this column to create and compute a Moneyline for each series match-up.

First I adjusted each team's expected Win % for Home-Field advantage and, conversely Away-Field disadvantage. It may be a few percentage points off so far this year, but it's been my experience that Home teams in baseball have about a 10% advantage playing at Home. To be conservative, I chose to adjust by 8% your numbers from column H. I increased the Home teams by 8% and I decreased the Away teams by 8%. I then created a Moneyline from my adjusted numbers.

How did I do this? You take the larger number and divide it by the smaller number.

This gave me a projected MoneyLine for each series.
You could do this on a day-by-day basis if you wanted to do the work involved in updating your spreadsheet and doing the math that I described above. I have been using the numbers from your spreadsheet that included games through May 18th. Last week, all 30 teams began a new 2-game series on Wednesday, May 19th, so I started there. I then continued, after I liked the results that I saw, with this week-end's games.

I then compared MY projected MoneyLine on all the games to the Closing Lines for all of the games and I charted the overlays (which are the picks). I found that when I had at least a 33 cent overlay on a game, it LOST!!
It lost VERY CONSISTENTLY. It lost 11 of 13 since Wednesday with the only 2 winners being Florida on May 19th and the Mets yesterday.

This is the idea that I mentioned in Post #721 and the picks that I'll be playing (small) today in Post #728.

~ Vinny Vidivicci

[Vinny Vidivicci]

For the sake of keeping things simple, let's name these 2 things that I have added to this thread. Since I added them, I'll name them.

I'll call the first one the "Window-8".

This second idea let's call "Reverse-Expectation".

I suggest you name what you have been doing Thurber as well, just to keep any tailers here clear on what they're reading or following. (just a thought)

~ Vinny Vidivicci

[therber2]

Sunday Window 8 Picks -


Washington
Houston
Oakland
Arizona +1.5 RL
Mets +1.5 RL

Sometime today I am going to post a simplified explanation and an easier way to compartmentalize and evaluate the picks on the Window-8 System. I will include year-to-date records that will simplify which parts of the system are of value.

The idea that I'm really excited about right now gave me the picks that I listed in post #728 which has a record of 11-2 since last Wednesday. I'll put a more detailed explanation of this on my next post after I've gotten my bets in at MatchBook.

~ Vinny Vidivicci

Vinny that would be great. I sent this to sycoog a bit earlier. I noticed the gap in your catagories being that you are missing picks on favorites. As you said before, you prefer to throw all of the seeds out there and hope that most of them grow (or something like that ). So I searched for trends on situations for favorites (a little deviation from the 8 day window):

Play on Favorites over -170 -1.5 RL for even odds roughly where it is game three of the series, the series is tied, and the favorite is 5 or more wins in their last 8 games for home if they are home or away if they are away (this is a gray area, because I noticed that home and away isn't all that relevant in this case) while the opponent is 5 or more losses (not sure if this part is needed).