What are ways you evaluate players that don't rely on analytics?

[giordunk]

What are some ways you evaluate players where you don't really rely on analytics, or things you think analytics can't measure?

I think there are some things like cultural impact that I will use to evaluate a player.

At the end of the day the NBA is also a business, and there is the aspect of basketball that evokes emotions and made me fall in love with the game in the first place, so I think it's reasonable for that to be considered in player evaluation, especially in a tiebreaker situation.

For example, I think Iverson has a higher place in NBA history than say a John Havlicek or Bob Cousy for what he meant to the game. He inspired a whole generation of basketball players across the world, including many that became NBA players.

A guy like Yao I also give a lot of credit too because of how much he grew the game.

[70sFan]

I try to eliminate fun factor in evaluating process, but at the same time I am aware of many of my biases. I don't think it's wrong to consider things like popularity or cultural impact as long as you state it clearly.

I don't think I agree that Allen Iverson was more important for the game of basketball than Bob Cousy by the way.

[MyUniBroDavis]

[SNPA]

Lol.

How could we possibly understand the game without a spreadsheet? If only there was a way to visualize the stats in human form as a performance.

[Jaivl]

An eyetest devoid of analysis is worth nothing though.

[WintaSoldier1]

-On Court Comfort/Poise (How relaxed/Prepared are they for a given scenario)

-Psyche; How the opposing team & the players own team treats the said Player.

-Motor; How often/When are they engaged in plays

-Shot Selection; What type of shots is this person taking? And what shots are they making?

Short list on some of things I look at

[MyUniBroDavis]

An eyetest devoid of analysis is worth nothing though.

Let me cook

[MyUniBroDavis]

^statistics and eye test are good to inform each other, going all in on one without the other if both data are readily available is stupid

Data can inform our eye test and tell us what to look out for, and also show us things we can’t see because we can’t see every game every team plays.

At the same time, yeah sometimes whether it be noise or situational people that create their all time peaks list from a spreadsheet aren’t exactly doing much more than those who watch a bunch of games and hate analytics. Two sides of the same coin.

[Owly]

What are some ways you evaluate players where you don't really rely on analytics, or things you think analytics can't measure?

I think there are some things like cultural impact that I will use to evaluate a player.

At the end of the day the NBA is also a business, and there is the aspect of basketball that evokes emotions and made me fall in love with the game in the first place, so I think it's reasonable for that to be considered in player evaluation, especially in a tiebreaker situation.

For example, I think Iverson has a higher place in NBA history than say a John Havlicek or Bob Cousy for what he meant to the game. He inspired a whole generation of basketball players across the world, including many that became NBA players.

A guy like Yao I also give a lot of credit too because of how much he grew the game.

Cousy is a curious choice as a counterpoint to an icon "because of what he meant to the game". I'm on the skeptical side on Cousy but for some time he was (otoh) hovering on around 10th on old fashioned "pre-analytics" rankings and whether explicit or not I think flash and narrative significance (perhaps not backed by significant driving force contribution to probably the primary narrative factor - titles) was driving a very high ranking (perhaps "normal" size [and whiteness??] helped in terms of generating hype around him at the time too?).


Whilst I have long pointed to Davies or Haynes, in the public eye Cousy was regarded as the original innovative ball-handler (and the prototype point guard too).


I would tend to think of "evaluating players" in terms of player goodness, though "evaluate" doesn't have to connote that. Fwiw there probably are analytics that would be a (crude) proxy for generating interest such as jersey sales, television ratings though that might significantly be influenced by availability and I guess perhaps you were more thinking of "APBRmetric", player goodness/production/impact measurement type analytics. Don't think it's wrong to evaluate/rank on generating emotion or aesthetics though that list is probably more a personal one and should I think be separate from a pure player goodness discussion.

I would quibble with the use of the phrase "has a higher place in history" as used - granting it is preceded by "I think" but that phrasing still seems off, makes it seem objective and general rather than incredibly fuzzy/vague and thus more personal. And the players in question were on a lot of title teams.

[Doctor MJ]

What are some ways you evaluate players where you don't really rely on analytics, or things you think analytics can't measure?

I think there are some things like cultural impact that I will use to evaluate a player.

At the end of the day the NBA is also a business, and there is the aspect of basketball that evokes emotions and made me fall in love with the game in the first place, so I think it's reasonable for that to be considered in player evaluation, especially in a tiebreaker situation.

For example, I think Iverson has a higher place in NBA history than say a John Havlicek or Bob Cousy for what he meant to the game. He inspired a whole generation of basketball players across the world, including many that became NBA players.

A guy like Yao I also give a lot of credit too because of how much he grew the game.

So, I want to distinguish between:

a) Methods of competitive basketball evaluation that don't rely on analytics.
b) Criteria which include non-competitive components.

For my methods, I tend to use analytics as sandwiched layer in between a qualitative start and a qualitative finish. I'm trying to find as much good analytical meat as I can to pack in there, but the bread that holds the thing together is always qualtative.

For my criteria it just depends on the specific question being asked. The Top 100 project is a competitive, career-based analysis. The Peaks project is a competitive, peak-based analysis. A Hall of Fame-type project includes factors that in and of themselves aren't about being their opponents.

[LukaTheGOAT]

You can take notes on a player's tendencies that stick out.

Noting how frequently a player misses an open pass, is probably the biggest one.

Also, counting how often someone gets blown by on the perimeter or overpowered in the paint, are usually things that stick out to me.

[MyUniBroDavis]

Lol.

How could we possibly understand the game without a spreadsheet? If only there was a way to visualize the stats in human form as a performance.

I’m confused if ur supporting what I’m saying or going against it lol

[penbeast0]

Analytics are one factor, and while I'm a bit behind on some of the impact and tracking stats, I try to mix them into the pot with all the other, more traditional factors. I trust defensive stats less than offensive ones so I tend to put greater value in (a) eye test, and (b) team performance for defensive anchors (mainly shot blocking bigs).

I also tend to count off court issues that I think would affect team cohesion and unity. Stars who party late before games (Barkley, Webber, etc.), guys who quit on teams before their contract is up (Anthony, Harden, etc.), players who publicly blame others or pick fights with teammates (Jordan, Green), players who fight substance abuse issues (King, Thompson). All those things tend to make me downgrade a player from his current impact because they wear on teammates and coaches. An unhappy workplace is not something I think is conducive to players playing their best, any more than a workplace where stars aren't accountable for playing both sides of the ball (effort, not necessarily raw talent).

[SNPA]

Of three items:

Eye test
Psychological profile
Advanced stats (anything beyond a boxscore)

…the advanced stats are worth the least.

[SilentA]

Teammate quality and/or coachability intangibles are the main one, based on whatever reporting or player/coach anecdotes we can find. At the end of the day, all time fantasy rankings or whatnot boils down to "would I draft them myself as a theoretical GM" to me. Something like this gives Tim Duncan a #1 candidate case, for example, despite lower impact than MJ and Lebron. It's also why I take Jokic over Shaq, and part of why I'd easily take Isiah Thomas (a responsible and dedicated team player with leadership qualities, despite flaws after playing career) over James Harden and Kyrie Irving.

Me having a bias for playoff performance is also arguably a factor. I don't reward people for sucking in the RS, but ultimately you only need to get top 8 to enter the playoffs, but top 1 to win in the playoffs. This is an oversimplified explanation (you can win a weak conference and get a favorable finals matchup), and seeding is a factor, but ultimately I don't think poor RS performances matter too much as long as they make the playoffs, whereas everything matters in the playoffs.

Also, due to how many variables there are in basketball, eye test exists for me to contextualize stats and develop a more complete overall player profile since stats cannot tell a 100% full story by themselves (though they're great for eliminating bias or discovering things not obvious). I'm not going to pretend to have the expertise to understand every action on both ends like a top coach, but I'm aware of my limits and I think I understand enough to have a more reasonable opinion than a lot of internet randoms lmao

[Blame Rasho]

This is an imperfect way to measure players but how well do they measure among their peers at certain ages. Look at a player like Vince Carter, at his apex might have been his 3rd year in the league. After that point, how did the following seasons did he compare to his peers.

The other variables that is hard to quantify is the ability to make other players better and how much certain systems elevate certain players. I don’t mean in just numbers, but in impact. I think of the great Chris Paul and Deron Williams debate. In no world would I take Deron Williams over Chris Paul. I think one of the sneaky wtf moments of recent memory was when Paul made the Thunder somewhat competitive in his year there. I also look at how Sloan somehow someway made a team without Stockton and Malone a .500 team the year after.

[Ryoga Hibiki]

no stats at all or not adcanced analytics?

[One_and_Done]

Stuff like how much a player 'inspired' ppl or grew the game etc is irrelevant to their ability to impact winning on the court.

Other factors of relevance beyond an eye test can include award voting. It doesn't tell you how good each guy was, but it does tell you roughly how each guy was perceived at the time. The conceptual model for a guy to succeed is important too. You look at guys like Dr J and you can see how he'd succeed today, then you look at what made Mikan or Moses successful and you get the opposite feeling. If a player doesn't have an archetype to succeed, or if that archetype is an extreme outlier like Charles Barkley, then the chances are high the player won't translate.

[AdagioPace]

First I try not focus myself only on "Scoring" because it will always be the easiest thing that catches your attention given the nature of the sport. Then I ask myself (who's holding the ball the most?)

it's a good exercise for personal "visual" satisfaction, but the moment you try to compare a player to another player, the need for something more objective sets in. The range of eye-based evaluations is pretty limited (mainly personal use). Even if you don't want to quantify and be scientific, establishing categories, using logic is definitely a requisite even for a crude comparison.

[AdagioPace]

You can take notes on a player's tendencies that stick out.

Noting how frequently a player misses an open pass, is probably the biggest one.

Also, counting how often someone gets blown by on the perimeter or overpowered in the paint, are usually things that stick out to me.

and that's how analytics was born, therein lies the "problem" lol!
the moment you start counting there's no going back from that

our modern mind cannot function on "good, bad, yellow, green," evaluations alone, unless you go back to pre-neolitic time

( I like your first point: taking notes about tendencies. Our brain can tell if something happens often even if we don't know exactly the frequency)