OFFICIAL: RipCityTwo evaluation of players

[magnifier661]

I decided to make an excel evaluation of our current players. So I need the forum's help on deciding the values. Your input will be greatly appreciated.

PPG: How valuable is this measurement on a scale of 100 total points? Then as we vote on players.

Example: If PPG = 100, and player "x" is a 20, then "x" will be a 20/100 available.

I was thinking of using these values, and we can vote on their total value. These aren't "black and white" values. This would be your perception of the player and how you value certain abilities in the game.

1.) Scoring
2.) Making other players better
3.) Scoring Efficiency
4.) Clutch
5.) Defense
6.) Character (using a negative to positive value)
7.) Teamwork (using a negative to positive value)
8.) Leadership
9.) Heart
10.) Poise (using a negative to positive value)
11.) Salary (should be counted as a negative)

I think some of what I posted are worth more than others, so the values should change accordingly. I also think that certain attributes like character and teamwork can have negative values, when their character, poise and selfishness becomes a factor.

Anyway, I think this would be extremely fun so we can get a generalized forum view of the current players. Once the model is in place, we can use it to evaluate other players the forum discusses to obtain.

THANKS IN ADVANCED!

 

[magnifier661]

Here is my vote:

1.) Scoring (1-50)
2.) Making other players better (1-100)
3.) Scoring Efficiency (1-50)
4.) Clutch (1-200)
5.) Defense (1-100)
6.) Character (using a negative to positive value) - (-50/+50)
7.) Teamwork (using a negative to positive value) - (-100/+100)
8.) Leadership - (1-100)
9.) Heart - (1-100)
10.) Poise (using a negative to positive value) - (-50/+50)
11.) Salary (should be counted as a negative) - (-10 per million)

 

[magnifier661]

Here is my rating, based on my values.

 

[Strenuus]

you batum basher!!! haha.

the only thing i'd disagree with at all is that robin lopez should have the highest 'making other people better' ranking. Other than that, I agree with most of this.

 

[BlazerCaravan]

Are your data based on manual entry or based on formulas from stats?

I love this!

 

[PtldPlatypus]

Rather than having salary as a negative, why not have it as a separate evaluator? That is to say, define a player's value using the other ten categories, and then separately express that value proportionate to his salary.

My reasoning would be that a player's value to the team doesn't necessarily change simply because he is paid more. It seems silly to me to say that Wes means more to the team than LA does, simply because he makes $8M less. However, it is not at all unreasonable to say that we currently get more "bang for our buck" out of Wes than out of LA, for that exact same reason.

 

[BlazerCaravan]

Rather than having salary as a negative, why not have it as a separate evaluator? That is to say, define a player's value using the other ten categories, and then separately express that value proportionate to his salary.

My reasoning would be that a player's value to the team doesn't necessarily change simply because he is paid more. It seems silly to me to say that Wes means more to the team than LA does, simply because he makes $8M less. However, it is not at all unreasonable to say that we currently get more "bang for our buck" out of Wes than out of LA, for that exact same reason.

That's a great idea. "Value" as an absolute number, plus Value Normalized By Salary".

 

[BlazerCaravan]

I'd love to do judgmental stat valuation. For example:

Positive Numbers:
Points/Assists/Rebounds/Steals in losses are counted normal.
Points/Assists/Rebounds/Steals in wins are counted as double.
Points/Assists/Rebounds/Steals in the clutch are counted triple.

Negative Numbers:
TOs/Missed FTs/Fouls in wins are counted normal
TOs/Missed FTs/Fouls in losses are counted double
TOs/Missed FTs/Fouls in clutch time are counted triple

Total all the Positive and Negative numbers, and get an overall number, weighted for important performance.

 

[magnifier661]

Are your data based on manual entry or based on formulas from stats?

I love this!

I actually would rather this be a general consensus with all forum members. I will accumulate the average of all posters and get a general average.

 

[BlazerCaravan]

I actually would rather this be a general consensus with all forum members. I will accumulate the average of all posters and get a general average.

That's a cool idea too!

 

[magnifier661]

Rather than having salary as a negative, why not have it as a separate evaluator? That is to say, define a player's value using the other ten categories, and then separately express that value proportionate to his salary.

My reasoning would be that a player's value to the team doesn't necessarily change simply because he is paid more. It seems silly to me to say that Wes means more to the team than LA does, simply because he makes $8M less. However, it is not at all unreasonable to say that we currently get more "bang for our buck" out of Wes than out of LA, for that exact same reason.

That makes a ton of sense!

Then on a separate worksheet, we could have +/- for salary.

 

[magnifier661]

That's a cool idea too!

Yeah, once we agree on the values, then I will do a thread for each player with the votes. The forum members can give numbers, on their perception, of each player on our roster.

Then we will have a community average!

 

[andalusian]

I think we have seen pretty well how poorly this team played last year without LMA, there is no way that he is tied in the making others better with anyone other than maybe Dame on this team.

Other than this, there is another piece that is really missing - and that is the "fit in the coach system" - because this is super important (otherwise someone like Hawes would be a greater value to this team than Kaman for example, which I am pretty sure we now know is not the case). FWIW - someone like Batum is really missing points to his real value to this team, imho - because he is such an important cog in the machine on both sides of the floor and his ability to guard some people that no-one else on the team can while also running the offense and allowing Dame to act as a defacto 2-guard at times.

 

[magnifier661]

I think we have seen pretty well how poorly this team played last year without LMA, there is no way that he is tied in the making others better with anyone other than maybe Dame on this team.

Other than this, there is another piece that is really missing - and that is the "fit in the coach system" - because this is super important (otherwise someone like Hawes would be a greater value to this team than Kaman for example, which I am pretty sure we now know is not the case). FWIW - someone like Batum is really missing points to his real value to this team, imho - because he is such an important cog in the machine on both sides of the floor and his ability to guard some people that no-one else on the team can while also running the offense and allowing Dame to act as a defacto 2-guard at times.

I do like "Fit the coach's system". I will add this. And don't think the chart I compiled is the final brother. Each member will be able to vote for each player. I will add all the votes, then make an average for the community. It doesn't mean this is the "end all", just the general consensus of this forum's observation.

 

[BrianFromWA]

You know, you could set this up in an eigenvalue-based matrix with s-curve analytics so that you wouldn't have to calculate the 1-xxx values, but have "basic" input values.

For instance, on a scale of -9 (clutch shooting) to +9 (rebounding), which do you prefer? I think both are important, but a slight edge to clutch shooting, so I might give a -2. You do that for all combinations. And then you can set up s-curves for hard-data analysis, so that instead of assigning Dame a grade based on what you think his "clutch performance" is, you can have curves set that give values based on data (for instance, clutch eFG% and "clutch time" FGA. Or whatever).

I think I didn't explain that right.

Repped for the thought. I'd love to help out on something like this.

 

[magnifier661]

You know, you could set this up in an eigenvalue-based matrix with s-curve analytics so that you wouldn't have to calculate the 1-xxx values, but have "basic" input values.

For instance, on a scale of -9 (clutch shooting) to +9 (rebounding), which do you prefer? I think both are important, but a slight edge to clutch shooting, so I might give a -2. You do that for all combinations. And then you can set up s-curves for hard-data analysis, so that instead of assigning Dame a grade based on what you think his "clutch performance" is, you can have curves set that give values based on data (for instance, clutch eFG% and "clutch time" FGA. Or whatever).

I think I didn't explain that right.

Repped for the thought. I'd love to help out on something like this.

I would love you to help. I could always set this up in google.docs if its easier for you