Analyzing a Trade
Moderator: Doctor MJ
Analyzing a Trade
-
Chicag-Fro
- Pro Prospect
- Posts: 786
- And1: 1
- Joined: Oct 13, 2002
Analyzing a Trade
I am just starting to delve into advance stats. I was wondering how some of you more experienced stat-heads go about analyzing trades. One way is to just evaluate the players against each other (ORTG, DRTG, TS%, etc.). However, how do you analyze the trade in terms of the expected impact it will have on the teams? For instance if you are trading a starting SG for a starting PF, what would be the best way of analyzing the trade. It would seem that just comparing stats head to head in this scenario wouldn't be a good indicator because you are not simply inserting one player in for another.
Re: Analyzing a Trade
-
jambalaya
- Sixth Man
- Posts: 1,674
- And1: 289
- Joined: Feb 01, 2005
Re: Analyzing a Trade
I'd look at player level adjusted +/- and lineup +/-.
Player level RAPM offers a simple estimate of what impact a guy has had. Of course it may not be the same in a new context and possibly a somewhat different role but it is a starting point and then you analyze the new context & role and try to revise the estimate for those conditions.
Does the trade screw up their best performing big minute lineups or does it allow them to stop using mediocre or bad ones and give them a player with a better Adjusted +/- than the guy he will replace in the team's big minute lineups? If the players coming in & going out are different positions, how good is the guy who will backfill the position vacated by the guy leaving and how good is his sub individually and how good are their lineups? Lineup data is usually thin to very thin on minutes played for all but a few lineups and thus the lineup results may bounce around but it seems worth trying to check if the lineups tend to outperform the sum of individual average stats or underperform them.
It is also important to cover most or all of the basketball roles to have an effective elite team. Duplication is some cases can be fine, helpful but if it comes at the expense of significant deficiencies that may erase much of the other gain or exceed it. How will the changes affect the team's performances on usage, shooting, turnovers, rebounding and getting to the line and on defense? Is the player going to get the ball as much as they did elsewhere, in the spots they want to be in and at good times on the shot clock or will they face significant changes? How adaptable are they capable of being and willing to be? Will the top guys share and work it out or conflict?
The performance / cost ratio of players now and in the future is generally important though some players can be less than stellar on the value ratio and still be the right / necessary ingredient for a specific team.
Looking at and giving weight to defense fully as much as offense will set you apart from most.
If your focus is on winning deep in the playoffs it would seem to me pretty important to know how well the players have done against the best teams, in the playoffs and in the regular season. Some players outperform their averages against the best, while most slip back some. Amount of playoff experience can matter but folks very in much they think that generally matters. Looking at champs though, most champs in the last 30 years have had rosters with greater than league average experience.
Player level RAPM offers a simple estimate of what impact a guy has had. Of course it may not be the same in a new context and possibly a somewhat different role but it is a starting point and then you analyze the new context & role and try to revise the estimate for those conditions.
Does the trade screw up their best performing big minute lineups or does it allow them to stop using mediocre or bad ones and give them a player with a better Adjusted +/- than the guy he will replace in the team's big minute lineups? If the players coming in & going out are different positions, how good is the guy who will backfill the position vacated by the guy leaving and how good is his sub individually and how good are their lineups? Lineup data is usually thin to very thin on minutes played for all but a few lineups and thus the lineup results may bounce around but it seems worth trying to check if the lineups tend to outperform the sum of individual average stats or underperform them.
It is also important to cover most or all of the basketball roles to have an effective elite team. Duplication is some cases can be fine, helpful but if it comes at the expense of significant deficiencies that may erase much of the other gain or exceed it. How will the changes affect the team's performances on usage, shooting, turnovers, rebounding and getting to the line and on defense? Is the player going to get the ball as much as they did elsewhere, in the spots they want to be in and at good times on the shot clock or will they face significant changes? How adaptable are they capable of being and willing to be? Will the top guys share and work it out or conflict?
The performance / cost ratio of players now and in the future is generally important though some players can be less than stellar on the value ratio and still be the right / necessary ingredient for a specific team.
Looking at and giving weight to defense fully as much as offense will set you apart from most.
If your focus is on winning deep in the playoffs it would seem to me pretty important to know how well the players have done against the best teams, in the playoffs and in the regular season. Some players outperform their averages against the best, while most slip back some. Amount of playoff experience can matter but folks very in much they think that generally matters. Looking at champs though, most champs in the last 30 years have had rosters with greater than league average experience.
Return to Statistical Analysis