Analytics: Sloan Convention

This convention that talks about sports stats and stuff this week. I'd like to gauge all your opinions on it.

Baseball - It is a rigid sport that accumulates data where it is usually 1 person vs 1 thing (batting % vs the curve). It basically works during the regular season and in the post season, it isn't successful because the games become more random.

Basketball - I'm mixed. I can get like FG% from certain areas on the floor, +/- is useless, per 100 possessions is useless as well

Football - it's too much of a moving sport to use any of them. Like I heard this guy on the radio just giving stat after stat after stat on why Mariota is better than Winston. I sat there and basically said, "Dude, none of that shit matters". There are too many types of offense, defense, play calls, situation, etc. etc. etc. to say why 1 player is better than another.

Like Demarco Murray and Shady McCoy. Murray statistically had the better year so everyone is saying, "oh he's the gr8st zomg". He played with 5 1st round picks on the Oline, a QB who converts on 3rd down, and a game breaking WR. He basically had to skip through some of those holes.

McCoy had no offensive line, a funky offense, and no QB or WR, still got over 1000 yards and another RB taking carries away. I think McCoy did more than his stats said. But analytic say he is bad because he ypc is down.

The only way I can it being useful in football is conversion percentages.

You're at you're own 45 yard line, 4th and 1. Do you go for it?

I think it would help answer some of those things. But like you said, football, there is just to much shit going on.

Never heard of this convention and I don't pay attention to stats. W's and L's is what matters. In fact there is way too many stats and there's only very few that actually matter. Red zone stats for one instance is important. TD to interception ratio is important. Turnover ratio for the team important. One of the problem is there are so many stats that could be achieved when the score is lopsided including passing yardage,completion % and YPG given up by the defense for example but they count and are used as arguments for why a certain player or team is better than another. For example Eric Decker was putting up nice stats for the Broncos but now is being dissed for putting up lesser numbers for the Jets. Well who's throwing to him? Did he just become worse by himself. lol.

Also the Eagles have a very good OLine. Maybe it's stats thats keeping them down like maybe a sack stat which could be cuz the QB held the ball too long. Remember the Sanchize.

Willbacker wrote:Never heard of this convention and I don't pay attention to stats. W's and L's is what matters. In fact there is way too many stats and there's only very few that actually matter. Red zone stats for one instance is important. TD to interception ratio is important. Turnover ratio for the team important. One of the problem is there are so many stats that could be achieved when the score is lopsided including passing yardage,completion % and YPG given up by the defense for example but they count and are used as arguments for why a certain player or team is better than another. For example Eric Decker was putting up nice stats for the Broncos but now is being dissed for putting up lesser numbers for the Jets. Well who's throwing to him? Did he just become worse by himself. lol.

Also the Eagles have a very good OLine. Maybe it's stats thats keeping them down like maybe a sack stat which could be cuz the QB held the ball too long. Remember the Sanchize.

Dude, the eagles Oline was shit. You had your one guy out the 1st 4 games, lost your center, both guards, and a couple other guys. Peters was the only guy who started 16 games I think.

I'll disagree with the basketball one. Per 100 possessions isn't important, that's for sure. It makes bench players look like studs in some examples. But +/- is important in terms of defense and offense that can't be tracked with any stats. You can have a lockdown defensive player that doesn't record many steals or blocks, but creates enough chaos and contests shots all the time, they're going to make a difference. Other stats like usage % are cool to look at to see how much a player really is involved in each player. I feel like usage should be tracked along side other stats like PPG or the +/- to really show how big an impact each player makes.

Basketball is a flow game. per 100 possessions don't show who the hell the player is up against. The usage % would really on count for PGs and shit. There was a stat that counted dribbles, that might be the closest to usage. Player X averages 17.77 dribbles per possession or what ever.

EPICxPENGUIN69 wrote:Basketball is a flow game. per 100 possessions don't show who the hell the player is up against. The usage % would really on count for PGs and shit. There was a stat that counted dribbles, that might be the closest to usage. Player X averages 17.77 dribbles per possession or what ever.

Ya but that is biased toward guards and smaller forwards. A Center or PF isn't going to put the ball on the floor as much

HurpDurpHurray wrote:

EPICxPENGUIN69 wrote:Basketball is a flow game. per 100 possessions don't show who the hell the player is up against. The usage % would really on count for PGs and shit. There was a stat that counted dribbles, that might be the closest to usage. Player X averages 17.77 dribbles per possession or what ever.

Ya but that is biased toward guards and smaller forwards. A Center or PF isn't going to put the ball on the floor as much

well duh. But you can't really track how long someone has the ball in their hands for every possession. You really want to be the guy with a stop watch in your hands? I mean conversely, who wants to be the guy counting dribbles. Which is why analytics is retarded.