Jump to content

Purpose of the Spread


The Wiz

Recommended Posts

I decided to give this topic it's own thread because I didn't want to clog up a game thread spread with this  discussion.

People use the spread info in different ways. Some use it for betting purposes. While it can be used that way that isn't the primary purpose. For those that do,  there is a way to find the discrepancies between Vegas and my system (people who have used it,  win more times than not). ....the over or under estimation of teams....In the business world this would be known as the inefficiency of the markets. In other words,  the market of points is skewed due to bias. What is this bias?  Here's is an example....At the start of next season Duke , Ky and Ks will all be ranked. How can I know this? We don't really know the teams for sure next year  ...but people will vote on them ...and bet on them to win.  People move forward looking in a rear view mirror. Even if the data next year shows these teams to be down people will still bet on them...at least for awhile.  And it works the other way....if the numbers are good ...let's say  Neb is highly rated...people will ask ...you mean football don't you? My fav is the Bills 2011-12 season ...we had come off of a bad season....no one had much hope for the upcoming season except maybe a few on this board....I was mildly optimistic at the beginning but didn't expect any Dancing....As we moved into Dec...it suddenly looked like we were going to be pretty good...as we moved into Jan...it became an OMG...as the Bills suddenly rang up an A+....I declared that on the board and many people backed away ...saying we are good ...but not that good....as we rose in wins and standings everybody hopped on board...well almost everybody.  The system can spot trends early.   Even without a full data set it can spot trends with as few as 3 games. However perception is reality and perception is slow to change.  Perception  leads to bias. So one of the purposes is to squeeze out the  bias.....though this isn't the main purpose of the system either.

What about the point spreads? While you  can use them to predict wins  that isn't their main purpose either. Suppose I predicted on the board that the next roll of the dice was 7....and it came up 12...there would be a series of posts here...The Wiz doesn't know what he is doing...the computer model is screwed up...he wasn't even close. The point is that regardless of the rants it doesn't change the fact that the correct call is 7. The difference between Vegas and bball at the Fetz...is the dice are fixed variables...7 will always be the favored roll...With basketball you have moving variables....different players playing different minutes...player output varying from game to game ...injuries ...different venues...etc etc.....a giant moving target. That's why I prefer college basketball ....moving variables...with lots of data....as opposed to say a crap game where the variables are fixed. In bball you start with no data and ramp up quickly. As you gain more data you become more accurate (usually) So what does this mean?

The importance of the spread is that it is a measuring tool.  To measure how we are doing as a team.  So when I miss on a point spread....what it  says to me is... the Bills had a better (or worse) game than expected.  Not that it was a wrong guess but that the Bills exceeded or under performed. In the SEMO game I had the Bills  as a 17 pt fav....we won by 10...no biggie ...1st game , right?  Anytime you miss by 3-5 pts you are outside the normal boundaries.  In this case the Bills under performed. I ran the game the next day as though the Bills were going to play SEMO again that night (I frequently run games after they are over  to see what happens) and the Bills came up as a 14 pt fav.... a big difference for 1 game. The significance at this point of the season with only 38% of the data  is that we have had 3 games  (a trend) against weak teams and have under performed. It is concerning ....the next 5 games can turn us around or lock us in.  The reason 8 games is important is there is enough data and data connection points to make it difficult to change much...In other words , you are pretty much who you are after 8.

What does this mean for the SH game........well subtracting 3 points for the large spread variable ( see SH spread thread Nov 15...1212 PM) and subtracting another 3-5 pts as a normal variation...if we lose by less than 13...that would be a good indicator  that we are doing better and would start a new trend line ...2 more good games (exceed the spread by 3-5 pts) and we are rolling.

In summary, the purpose of the spread is that it is a measuring tool.  It can be used for a betting device or to predict wins but the main reason is to see how we are doing ..which way are we trending...how are we in comparison to other teams (grading system)....When I post i don't just say these are the facts....I try to use the data to show how we can win...or afterward ...how we lost.   I am fan of the Bills and college basketball.... I do this because these are the things I want to know as a fan and I post here because I assume other fans would be curious and find it interesting. I think this season has a lot of promise  but first we need some things to come together ....Hopefully, it will be sooner rather than later.  In any case , I will always let you know

 

Link to comment
Share on other sites

Appreciate the explanation, Wiz, but, always a but right, if performance thru game 8 defines a team why does the selection committee always look at how a team may have come on strong in Feb or early Mar as a reason to select an outlier? It may take a team like SLU more than 8 games to find out just who they are. I hope it doesn't, but it very well could. 

Also your spread against SEMO was pretty good. What model can predict 16% from the three and 50% from the line? 2 more 3's and your line's almost perfect. Tricky hobby you chose, but you've got nothing to prove to us. You are a Wiz.

All that said, I don't understand the SH by 21, but that's why they play the games, right? 

Link to comment
Share on other sites

4 minutes ago, slu72 said:

Appreciate the explanation, Wiz, but, always a but right, if performance thru game 8 defines a team why does the selection committee always look at how a team may have come on strong in Feb or early Mar as a reason to select an outlier? It may take a team like SLU more than 8 games to find out just who they are. I hope it doesn't, but it very well could. 

Also your spread against SEMO was pretty good. What model can predict 16% from the three and 50% from the line? 2 more 3's and your line's almost perfect. Tricky hobby you chose, but you've got nothing to prove to us. You are a Wiz.

All that said, I don't understand the SH by 21, but that's why they play the games, right? 

Right...that's why they play the games....

As for the final 10 games...part of it is a momentum indicator....You want a hot team and part of it is you are playing the best teams ...battling for 1st ...plus playing  in your conf tourney.. ...if you are playing for a Dance spot...you should be hot

Link to comment
Share on other sites

Appreciate the explanation Wiz, and of course, from the point of betting, you really know a whole lot. However, I am more interested in the use of data to determine issues like over rating and under rating. Just for confirmation, are you saying you use the deviation to the spread as the basis for these determinations?

 

Link to comment
Share on other sites

Wiz, you do you. If posters take issue with how you calculate their spreads, they can choose not to read. You’ve done this for years, and I think your efforts are valuable and fun. There’s a reason no other poster does what you do. Nobody beats the Wiz.

if you’re wrong, so be it. Your model will strengthen from the extra game. Thanks for continually doing this, and any body who complains can try to do better I guess. My bet is they can’t.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...