APBRmetrics

devin3807 · Joined: 11 Oct 2007 Posts: 66

Ok, you all know the phrase "the first team to 100 wins".

I heard it recently during a game (I don't remember which), and I thought that it would be interesting to actually see the numbers on this, and at various other points.

Intuitively it makes sense:

-If you're the first to 100, that obviously means that you're in the lead.
-Games usually end around 100 points.
-If you're in the lead near the end of the game, you've probably got a better chance of winning, as a opposed to "the first one to 25", where you're obviously in the lead, but likely with about 3/4 of the game left to play and plenty of time for things to change.

Anyhow, I'd be interested in seeing some numbers on it.

Is there anyhow (website?) to punch in the numbers like this to yield the final record under these conditions? Ex. First to 100, first to 104, first to 60, etc.

Further, would it be reasonable to assume a steady climb in correlation between how high the score is, and how often they win?

For example, as an aggregate, would you assume that the first to 101 (and winning) actually has a higher correlation than first to 100, and then first to 102 (and winning) actually has a higher correlation than first to 101, and so on.

I would assume so, but I'd be interested to hear other views.

Ryan J. Parker · Joined: 23 Mar 2007 Posts: 711 Location: Raleigh, NC

I'm not aware of anything that exists where you can easily query for this sort of stuff, but you could use the data available at basketballvalue.com to calculate these sort of things.

Kevin Pelton · Posted: Tue Apr 29, 2008 1:38 pm Post subject:

First to 100 is also known as "Lawler's Law," as coined by long-time Clippers play-by-play guy Ralph Lawler.

According to Lawler's Wikipedia page, from 1978 through some time, Lawler's Law held 91.5% of the time in Clippers games. I came across an old blog of mine where I referenced an ESPN graphic saying it held 94% of the time over the past five years (this was cited early in the 2005-06 season).

S.K. · Joined: 18 Feb 2005 Posts: 61 Location: Toronto

This is pretty obvious, isn't it? The first team to X (where X is any number > zero) is more likely to win than their counterpart (since it's better to be winning than losing), and the higher X is, the greater the percentage (since by definition it'll be closer to the end of the game).

Let me rephrase this: teams that are winning late in the game are likely to win.
_________________
No books - no articles - no website.
Just opinions.

Ill-informed opinions.

Mike G · Posted: Thu May 01, 2008 1:58 pm Post subject:

I guess the 'first to 100' situation is reached about 60% of the time -- the other 40%, neither team reaches 100 -- and 90% of those times, it turns out to be winning team. So it's more graspable than "first team to 130 wins 99% of the time", or "first team to 40 wins 60% of the time", etc.

It therefore seems to have a little mental exercise built in, for those who don't do such things regularly. I doubt many coaches or players have improved their winning% by steering toward such a benchmark. It's just something to say because you can.
_________________
`
36% of all statistics are wrong

gabefarkas · Joined: 31 Dec 2004 Posts: 1313 Location: Durham, NC

mtamada · Joined: 28 Jan 2005 Posts: 377

Mike G · Posted: Wed May 07, 2008 7:56 am Post subject: