System Plays 1/22

[Cubbiefan16]

Current Record: 14-13 (51.9%)


On a good run recently, 6 for the last 8. Hopefully it can continue today. Three plays today, but none of which are very strong. Here they are:


Dayton -5

Louisiana Tech +7.5

Weber St. -11



Like I said, none of those are of great strength. Let's see how they hold up. Good luck today everyone

[hartzell]

I've noticed in my system, which of the past 56 games it has picked is only 55%, that when a team is favored and on the road, they are 10-1. I'm not too sure why, but all the others are around 50%. Hopefully this is something that will hold. So with that said, I like Dayton at 5 and I've also got UCLA at -5.

Good luck tonight.

[superfuzz]

I like the La Tech play. GL tonight!

[Cubbiefan16]

hartzell - That is interesting - trying to figure out a reason why that would be. How is home court advantage factored into your system, if at all? My home faves have hit at a better percentage than any other..

Also, UCLA is very nearly a play today in mine too. About as close as it comes without making it. Hope they both hit.


Thanks superfuzz, let's get it. Good luck to you too.

[hartzell]

So far, the only the only thing that accounts for home court advantage is average points scored at home vs the other teams points scored away.

Also, I like the other two picks. I have them both covering but they don't follow the current pattern I've been working. I even have La Tech winning, but I'm weary when my system picks dogs to win.

Regardless, Lets get them tonight!

[Cubbiefan16]

Originally Posted by hartzell

So far, the only the only thing that accounts for home court advantage is average points scored at home vs the other teams points scored away.

Also, I like the other two picks. I have them both covering but they don't follow the current pattern I've been working. I even have La Tech winning, but I'm weary when my system picks dogs to win.

Regardless, Lets get them tonight!

The problem I see with that is one, strength of schedule isn't included. A team is going to score more points, regardless of home or away, if they're playing a NJIT or Alcorn St for example, compared to a team playing a defense like Duke. Secondly, if a team has played in slow paced games against say Iowa or Oregon St., they will automatically have fewer points scored than a team playing Wake Forest or Texas Tech, again, regardless of how good their offense is or how they play on the road. For me, this was one of the main problems in my previous systems - I had too many road dogs and favorites because home court wasn't factored in enough. But it's also inadequate to just assign a certain value to add to home teams and take away from away teams -- because certain teams play better on the road, certain teams have bigger home court advantages, and the paces, etc, could be different, etc, etc. That's where I think my main problem was before. That's why, now, I use home court as a base in my system and almost doubly factor it in. In addition to it being the base of the computer's plays, I also integrate it in using efficiencies. I put each team's home/away offense/defense efficiencies into the model and this helps to adjust for home court advantage even better. In this way, pace is standardized as well as SOS being included. I use my own formula for calculating efficiencies (Statistics Major), but if you're unfamiliar with this, you can always use an outside source (KenPom is usually pretty good and there's another one too, but I can't think of the name right now..). I would get into more of what I do, but I think this is starting to get a little boring. Sorry again for the novel and if I'm giving you more information than you want.. Good luck tonight -- hope our matched plays work well...

[mcallister3200]

GL cubbie, nice to see it turning around for ya!!!

[hartzell]

Thanks for the reply and don't worry about the novel as I am about to write an essay myself. This kind problem solving is interesting to me so you can talk my ear off all you want. My basis for the home court advantage was mainly on points, but it takes in to account the difference between playing at home or on the road. In my calculations, home team only gets home statistics and vice versa. I also use the strength of schedule and RPI ratings. The problem is that I'm not sure I'm applying those last two factors correctly. I've also become torn on how strong I take other factors in relation to others. Assuming my system was complete, which it is far from, the only thing that stuck out as a consistent pattern was those road faves that also agreed with my system. Home faves/dogs and road dogs were all pretty equal in number in winning vs losing. I haven't had the time to go back and recalculate for all of the past games. But enough of that, I have most of tomorrow off which I will spend retooling my system. I hope I can figure this out soon to stop the bleeding.