How often do people use the eye test as a cop out argument?

[atljonesbro]

It seems like whenever facts dont fit a posters agenda they'll quickly denounce them and shout eye test because you don't have to back what you say. Are people just getting lazy or so they truly believe they don't need to back what they say. Can I say Al Horford is the best center in the league, the eye test says so?

[Clyde]

It seems like whenever facts dont fit a posters agenda they'll quickly denounce them and shout eye test because you don't have to back what you say. Are people just getting lazy or so they truly believe they don't need to back what they say. Can I say Al Horford is the best center in the league, the eye test says so?

I'm a huge Gator fan, but you need to get your eyes checked if you're serious.

[atljonesbro]

I'm a huge Gator fan, but you need to get your eyes checked if you're serious.
Just using an example of some trends I see around the forum in terms of how people present their eye test argument

[The Repo Man]

It seems like whenever facts dont fit a posters agenda they'll quickly denounce them and shout eye test because you don't have to back what you say. Are people just getting lazy or so they truly believe they don't need to back what they say. Can I say Al Horford is the best center in the league, the eye test says so?
I dunno it's hard to estimate. I'll say maybe at a 40% clip.

As in 2/5.

[Clyde]

The eye test only counts if you know what you're looking at/for.

[SOD 21]

Why do stat geeks try to exclusively use advanced metrics to make their arguments and completely ignore what they see? Shouldn't it be a combination of both?

Stats and advanced metrics are only part of it.

[atljonesbro]

I dunno it's hard to estimate. I'll say maybe at a 40% clip.

As in 2/5.
You seem pretty insecure. This thread isn't about LeBron.

[atljonesbro]

Why do stat geeks try to exclusively use advanced metrics to make their arguments and completely ignore what they see? Shouldn't it be a combination of both?

Stats and advanced metrics are only part of it.
This thread has nothing to so with what you're talking about.its about how people will straight spew garbage and say they're right because the eye test says so

[TheGreatDeraj]

I can't speak for other people, but when I see statistics that do not match the impact that I see on the court, then the "eye test" is more valuable to me.

I still use statistics, and they give you a general idea of what was produced on the court, but the impact on the court cannot be determined without watching the game and using context.

[GODbe]

If you're not smart enough to be able to judge a player with your own 2 eyes you don't deserve to live. It's that simple.

Eye test > stat nerds

[Inactive]

Why do stat geeks try to exclusively use advanced metrics to make their arguments and completely ignore what they see? Shouldn't it be a combination of both?

Stats and advanced metrics are only part of it.I don't think I've seen many people argue solely on the basis of advanced metrics. Mostly people use metrics to support an argument which is based on what they've seen, or to refute an argument which they oppose.

When people bring up the eye test, it's usually just a way of dismissing evidence contrary to their opinion. If they posted game footage, or talked about the actual situations in game that they were referring to, then it would be perfectly fine.

The conflict isn't really between stats, and the eye test. It's between arguments based on verifiable evidence, and arguments based on the flapping of one's gums.

[fpliii]

Well, there are three phases to my evaluation, normally:

1) Figure out what kind of role a guy played. Rate adjusted box score stats, and watching games (i.e., the eye test) are important here.

2) Look at metrics such as RAPM (since 96-97, not available earlier since the play-by-play data does not exist), or with/without, and team ORtg/DRtg (NOT individual ORtg/DRtg numbers, those are useless and based on box score numbers), to try and determine how effective a player is in his role. Useless shit based on box score metrics (I don't look at the aforementioned ORtg/DRtg, win shares, WS/48, DWS, OWS, PER at all, they don't tell me anything that I can't tell from adjusted box score stats).

3) Watch more games (eye test again) and determine *how* exactly a guy does what he does. In addition, I'll look at shot charts (and splits against good defenses), Synergy data, read descriptions/breakdowns/analysis by other players or posters on this board. For the most part here, I want to get a very good idea of a guy's skillset. In particular, I want to make sure a guy's game translates to the playoffs (and for older guys, I need to see things in their games that will convince me that their play would translate to this era...if it doesn't, I have to try and decide whether the deficiencies can be easily fixed, or if the player would be unable to adjust to today's game).

A bit of a rant, and it's not nearly as rigid as it sounds. Just trying to describe the general mechanism by which I compare and evaluate players. Maybe I'm going about it the wrong way, but I feel like I get a good idea of what's going on by doing this. :confusedshrug:

[ralph_i_el]

It seems like whenever facts dont fit a posters agenda they'll quickly denounce them and shout eye test because you don't have to back what you say. Are people just getting lazy or so they truly believe they don't need to back what they say. Can I say Al Horford is the best center in the league, the eye test says so?

eye test is necessary to evaluate the context of a players stats. You have to understand what role that player has, how many touches they get, how many plays are run to get them open. You have to watch the games to see how defenders guard them or give them space. Defensive metrics are nice, but you can post great metrics if you never guard the opponents good offensive big/win/guard and are always hidden on weaker players (Kobe rarely guards the opponents best wing/guard in recent years for example).

more examples that you need to watch games for:

Does the player get tired and drop off?

Do they beat their defender with quickness/strength/ or savvy

Can they make tough shots?

How long is the set-up required to get them a good look? What kind of plays do you run for them?

Do they move the ball quickly? Some guys get lots of assists but pound the rock and don't pass as much unless their is an assist oppourtunity. Other guys move the ball quickly and get more secondary assists that don't show up on the stat sheet.

DEFENSE?

Quick or slow release on the shot?

Are they consistent or Hot/Cold?

Drive to shoot or drive to pass? Do they like to pull-up off the drive?

Can they play within their skill set? Or do they constantly try to do too much (LOOKING AT YOU JAVALE)

Do they look for their jumper (curry) of only shoot when given space (Griffin)?



It's more important to know what roles a player can fill and what tools they have, rather than measuring their production.

[atljonesbro]

eye test is necessary to evaluate the context of a players stats. You have to understand what role that player has, how many touches they get, how many plays are run to get them open. You have to watch the games to see how defenders guard them or give them space. Defensive metrics are nice, but you can post great metrics if you never guard the opponents good offensive big/win/guard and are always hidden on weaker players (Kobe rarely guards the opponents best wing/guard in recent years for example).

It's more important to know what roles a player can fill and what tools they have, rather than measuring their production.
I'm not asking why we use the eye test, I'm asking why so many people use it as a cop out argument when nothing fits their agenda.

[TheGreatDeraj]

I'm not asking why we use the eye test, I'm asking why so many people use it as a cop out argument when nothing fits their agenda.

an example? On ISH I'd say it's 90% trolls, so that would probably explain it...

[I.R.Beast]

I can't speak for other people, but when I see statistics that do not match the impact that I see on the court, then the "eye test" is more valuable to me.

I still use statistics, and they give you a general idea of what was produced on the court, but the impact on the court cannot be determined without watching the game and using context.

this... I apologize for bringing LeBron into it... but James has only impacted 1 game in the finals...and that's the game they won... All the other games he essentially padded stats with the game out of hand...

[Levity]

I'm not asking why we use the eye test, I'm asking why so many people use it as a cop out argument when nothing fits their agenda.

maybe use said Eye Test and see who you're actually conversing with in here. cause chances are, like its already been mentioned, youre getting trolled.

[Myth]

Kobe fans are the only ones to rely on the "eye test" argument. Everybody uses it as part of their evaluation process, but no other fans rely on it so much. The dumb part is, the eye test shows that Kobe has always had bad shot selection.

[BoutPractice]

fpliii > What exactly do you mean by "adjusted box score stats"? Just curious.

Anyway, the answer is often. Stats don't tell you everything you need to know, but when someone just invokes "the eye test" and doesn't even elaborate (or at least doesn't make it clear what they're talking about) it's hard to take their argument seriously. (Not that I haven't done it myself sometimes, but then again I don't take my own arguments that seriously...)

If truly your "eye test" tells you something important, then you should be able to describe in detail what you see and why it matters. A post by someone like kblaze, for example, doesn't primarily argue based on stats, but is actually full of relevant observations about the game.

[fpliii]

fpliii > What exactly do you mean by "adjusted box score stats"? Just curious.

Anyway, the answer is often. Stats don't tell you everything you need to know, but when someone just invokes "the eye test" and doesn't even elaborate (or at least doesn't make it clear what they're talking about) it's hard to take their argument seriously. (Not that I haven't done it myself sometimes, but then again I don't take my own arguments that seriously...)

If truly your "eye test" tells you something important, then you should be able to describe in detail what you see and why it matters. A post by someone like kblaze, for example, doesn't primarily argue based on stats, but is actually full of relevant observations about the game.
TS% (with true shooting attempts, FGA+.44*FTA), TRB%, AST%, TOV%, etc.

The exact numbers don't matter, but I just want to eyeball so I can get an idea of what kind of role a player has on his team.

[BoutPractice]

Thanks. Yeah, that seems to make sense.

[SouBeachTalents]

The eye test imo can be applied to specific games or series, like if a players final numbers look better than what that player's actual impact was during the game/series.

For example, LeBron's triple double in Game 5 of the 2011 Finals might look like a pretty good performance if you only looked at the box score, but if you watched the game you could see LeBron did not have the impact on the game that those numbers would indicate

[TheGreatDeraj]

Kobe fans are the only ones to rely on the "eye test" argument. Everybody uses it as part of their evaluation process, but no other fans rely on it so much. The dumb part is, the eye test shows that Kobe has always had bad shot selection.

Didn't want to go there, but:

sometimes kobe does make bad decision, but it's a myth he has bad shot selection

most of his "bad shots" are
-against a mismatch
-shots coming as the double is coming(shooting as the defensive scrambles opens up offensive rebound because the defensive is in scramble mode, it's harder to boxout, and it's a 4vs3 rebounding edge for his teamates if you try to double Kobe and he shoots)
-his "bad shot" is taken with his bigs in position to get offensive rebounds
-the "bad shot" is a midrange that breaks the defense because the defensive gameplan is to stop the paint and the 3, so it is a killer to make shots the the defensive is "giving"
-he's feeling it

for example:
they may look like bad shots, but a 45% midrange shot with a 25% chance offensive rebound a lot of the time is better than a 35% 3 point shot with 5% chance of offensive rebound

[Inactive]

The eye test imo can be applied to specific games or series, like if a players final numbers look better than what that player's actual impact was during the game/series.

For example, LeBron's triple double in Game 5 of the 2011 Finals might look like a pretty good performance if you only looked at the box score, but if you watched the game you could see LeBron did not have the impact on the game that those numbers would indicateThis is nonsense. In what way were his points, rebounds, and assists less valuable than any other player's? His stats weren't great that game, but they were no less valuable than the same stats on any other night.



for example:
they may look like bad shots, but a 45% midrange shot with a 25% chance offensive rebound a lot of the time is better than a 35% 3 point shot with 5% chance of offensive reboundIs there any evidence that ORB% is lower on 3 point FGA?

[TheGreatDeraj]

This is nonsense. In what way were his points, rebounds, and assists less valuable than any other player's? His stats weren't great that game, but they were no less valuable than the same stats on any other night.

Wrong. When the offense is gong great and you are getting defensive breakdowns you can often get any shot you want. These are easier shots to get and make.

However, when the defense tightens up and the offense is NOT going well, and the other players on your team are not getting good looks, those points are more valuable because they are more difficult

for example: is it more valuable to score 2 points early early in the game when your team is scoring easily against the defense or how about later in the game when you are down by 10 trying to come back and the defense is tightened up? which points are more valuable? Which adds more impact to the game? Which helps your team out more?

[fpliii]

Is there any evidence that ORB% is lower on 3 point FGA?
http://courtvisionanalytics.com/where-do-rebounds-go/

It's lower on 3PA than shots at the rim, but midrange shots yield a comparable or lower ORB% than threes do.

[TheGreatDeraj]

This is nonsense. In what way were his points, rebounds, and assists less valuable than any other player's? His stats weren't great that game, but they were no less valuable than the same stats on any other night.

Is there any evidence that ORB% is lower on 3 point FGA?

the rates are going to change depending on the players on the court and their positions.

Longer shots generally have longer rebounds. on three point shots you are more likely to see the tip rebound where the big guy bats it out to a guard, or a guard or wing gets it directly

closer shots generally have shorter rebounds and are gathered more often than not in the paint by big guys

[TheGreatDeraj]

http://courtvisionanalytics.com/where-do-rebounds-go/

It's lower on 3PA than shots at the rim, but midrange shots yield a comparable or lower ORB% than threes do.

that is in general for the NBA that is not based on the individual teams gameplan. A drive and kick offense will get LESS rebounds on midrange shots than a post play based offense, so you have to look at it play by play depending on what the team in running, not on overall statistics

nice link I'm checking it out now, have never seen that

[fpliii]

that is in general for the NBA that is not based on the individual teams gameplan. A drive and kick offense will get LESS rebounds on midrange shots than a post play based offense, so you have to look at it play by play depending on what the team in running, not on overall statistics

nice link I'm checking it out now, have never seen that
Oh obviously, depends on positioning on team rebounding philosophy (some will box out and crash the boards, some will get back in transition to defend).

[Inactive]

Wrong. When the offense is gong great and you are getting defensive breakdowns you can often get any shot you want. These are easier shots to get and make.

However, when the defense tightens up and the offense is NOT going well, and the other players on your team are not getting good looks, those points are more valuable because they are more difficult

for example: is it more valuable to score 2 points early early in the game when your team is scoring easily against the defense or how about later in the game when you are down by 10 trying to come back and the defense is tightened up? which points are more valuable? Which adds more impact to the game? Which helps your team out more?They both have a 2 point impact. They help the team equally. The guy who broke down the defense to create that easy shot deserves some credit. Being able to make tough shots is valuable, being able to create for yourself is valuable, but points are points. Their value doesn't change.

[Graviton]

Westbrook is a good example of eye test vs box score watching, even though he puts up near triple doubles. He may have a low FG percentage at times but that won't tell you how many times he attacked and broke down the defense, how many fouls he drew, how many plays he made but didn't get an assist, how many of his misses resulted in offensive rebounds because 2 people were busy chasing him, how many times he took a bailout shot with seconds on the shotclock because the offense broke down.

The box score specialist will call him a chucker, yet the guy who watches the game would see him soaring for rebounds over big men, diving for loose balls even though it's only the 1st quarter, putting pressure on the defense and tiring out his matchup while opening up the floor for everyone else. A walking energizer bunny is what you will see, but the box score alone would never tell you that much.