Can someone explain to me how Kiko Alonzo had a "disappointing" season?

Assume they have trouble ID'ing stuff. That would apply to all their grades and not preferentially to the worst QB's. The problem we're seeing is primarily with the worst QB's so any explanation has to naturally apply more to the worst of the bunch than to the rest of them.

btw.. I'm not really surprised by what you see in those graphs. People in general have trouble remaining internally consistent when you ask them to rate things on a scale with too many possible rating categories. 0.0 to 20.0 or 0 to 100 (even if that's a sum of different scores) is just way too large for most people to remain internally consistent. It's precisely for this reason that questionnaires in most fields limit the number of rating categories to often only 5 or fewer rating categories (and at worst 10). And even 5 may be too much. Netflix went from a 5-point scale to a 2-point scale precisely because it was "too confusing" for customers (EDIT: looking this up.. looks like it really had to do with misinterpreting "average" rating for "suggested" rating, so maybe not the best example).

 

No, the issue is that the better offenses w/ presumably the better QBs are more likely to execute the play correctly which results in a positive gain one way or another; that situation makes for an easy grading assessment. The problem arises when the pass is incomplete, and those at PFF are incapable of correctly determining the reason for the failed play. IOW, they have troubles ID'ing who was responsible on the play, and are more likely to blame the QB, b/c that's what amateurs are more inclined to do. And that is more likely to happen with struggling offenses, and presumably lesser QBs.

 

Interesting! OK that does fit the data then.

Of course that would mean that whoever was responsible is not getting as much blame as he should be getting. So we can test this by looking at comparable distributions of WR ratings. WR ratings should NOT show the same behavior is what your hypothesis would predict. Correct?

 

That's assuming that PFF understands the route concepts run vs the coverage given and that the receiver is running the correct route or making the appropriate option adjustments vs that coverage.

 

Every situation is different and must be assessed separately. I'm not one to settle with general or blanket statements for the sake of just trying to make sense of it thru metrics. It can't be done in football with much accuracy.

On the other hand, tape doesn't lie.

 

Wait a second. Whether they understand that or not, they're assigning too much blame to the worst QB's. They can't simultaneously assign too much blame to everyone responsible and not have a bias. So extra blame assigned in one place MUST be compensated for by extra credit assigned someplace else IF there is no bias but just a lack of understanding.

So if we find a similar issue with WR rankings, then it's most likely due to bias rather than error.

I have to be real careful with WR's though. Basically, I need to weight the WR ranking by how often that WR is targeted and group WR's into units so that we have 32 data points. Otherwise we won't know whether credit gained/lost by one WR was compensated for by another on the team. It will take some time to get the data (I don't have WR databases handy as I have QB stats), but I think it's worth doing because you raise an interesting point that can be tested.

 

Tape doesn't lie but people do, or at least people often don't process information on tape correctly. That's why it's important to test things whenever possible because people often make mistakes even when they think they're not making mistakes.

 

Kudos. Hard to argue with this when our current discussion is about PFF misdiagnosing plays.

 

Some research has shown that ,given competent training to do a task, that the people most confident about not making mistakes actually made more mistakes than the least confident. Whether that is something that applied to the specific situation, or is more general needs following up.

 

I believe this. My anecdotal experiences have been the same. For example, I seemed to always do better on a test when I thought I did bad. And the opposite was also true. I'd finish a test and be like, "I nailed it!" And end up with a "B". The next test I'd be like "Damn....fail" and I'd get 100% correct. It never failed and I couldn't help it. lol

 

My kid and the neighbor's kid is in band, and each year they have to audition for the region band and the all-state band. Every year, the neighbor kid says he will come in 1st place for his section- he thinks he's awesome (he's not- he's average for his age). Meanwhile, my kid comes out of the audition saying she did horrible, messed up a thousand things, etc. My kid is usually top-5 in the state. The neighbor kid usually isn't in the top 100 (he did just come in 15th this past weekend though, which was awesome for him).

Now, my kid practices a lot more than the neighbor, so I think work ethic has a lot to do with it. But the neighbor has a massive advantage in that he goes into these auditions with zero nervousness- he does his absolute best every time and it shows in his scores. Meanwhile, my kid is debatably the best high school sax player in the state but she's a nervous wreck when it's audition time- so she does worse than her talent level every time.

So while I think you're right overall, I think you're wrong on how important that confidence is. The neighbor kid is expecting to do well and overachieves, while my daughter expects to be perfect and underachieves. If my kid could have the neighbor's confidence, I think she would have been #1 this year (her teacher does too).

My kid can play her audition piece PERFECT every time EXCEPT in that darn audition...it drives me nuts I can't get her to fully believe in herself.

 

Key Carmen.. you might be right. PFF graders might be incapable of correctly diagnosing when the QB vs. receiver is at fault for the worst offenses. Here's the comparable graph to the 1st graph in post #39, but for WR units in 2015:


That's a linear relation there. So what that says is that PFF graders unfairly penalize QB's on the worst offenses (post #39) but not their WR units. That means there's something that either the WR or the defense (or another unit like OL etc..) is doing that is unfairly being blamed on the QB, and primarily for the worst offenses.

This alone doesn't answer the question of whether there's no bias whatsoever (would have to look at ratings for the defense and see if the extra blame on the QB is being assigned as credit to the defense), but it does support your suggestion that PFF graders can't properly diagnose who is at fault in certain cases.

For reference, I used the WR ratings from here in 2015:

then weighted each by number of targets and took the weighted average rating for each team, which is plotted above.

 

Nice job. Very nice. I began to notice their errors a few years back in terms of improperly ID'ing coverages, and if you can't get that right then it throws off everything else that you're trying to analysis.

 

My personal view on reading pff grades is that I trust them more on player groupings that are more individual performance and less on player groupings that depend on co-ordination. So DL, RB, WR I have no real gripes with. Their OL ratings I don’t trust at all, and I have low trust on their LB and QB ratings. Their DB ratings are more or less OK, but can spit out strange anomalies.

However at the end of the day pff ratings are not true statistics, they are merely a number attached to an opinion.

 

Here's another spot on analysis of Alonso from a very reputable source.

http://bleacherreport.com/articles/...he-top-4-3-outside-linebackers-of-2017-season

 

Put down the pipe.

 

I think most agree he "took a step back" this season. I so as much every time I talk about the guy. He started the season off strong, however. He didn't just "lose it". He was asked to do too much and he failed at it.

The frustrating thing, for me at least, is this Madden-esque way of looking at football. As if any team would ever be able to put all pro's at every position. It doesn't work that way.

Bottom line, he's a decent LB. Can we do better? Of course. You could say the same about 95% of the players in the league.

 

He's a liability and a detriment. There's no way around it. Obviously not every unit is going to be loaded, but competence should be expected.

Speaking of losing it, he hasn't had it since his rookie year in Buffalo. He's had 3 ACL surgeries (1 on one knee, 2 on the other), and a hip surgery. He's not the quick, agile LB he once was.

 

I don't think your injury history on him is correct. Are you confusing him with someone else? He's had a labrum tear, and one ACL grade 2 sprain and a grade 3 ACL tear on his right knee.

 

Nope. He injured is left ACL in college and with Philly, and his right ACL and also had hip surgery for a labrum tear while in Buffalo, too.

 

Nope. He never injured his ACL in college. In 2009 he played in 12 games. In 2010 he played in 1 game because he was suspended for DUI. In 2011 he played in 12 games. In 2012 he played in 12 games.

 

C'mon man, don't make me do your homework for you.

Give me a sec.

http://www.buffalobills.com/news/ar...ing-well/9f483467-b54c-410a-81db-8fd917871f14

http://bleacherreport.com/articles/...-things-you-need-to-know-about-the-oregon-ilb

 

Alonso missed 2010 from a DUI-related suspension while simultaneously rehabbing his torn ACL.

Oh my, I guess he did injure his ACL in college...number 1888 things that someone got wrong.

Now, back to why Kiko doesn't get love...

Making tackles 8-12 yards past the LOS doesn't really count as a tackle more than it counts as 8-12 yards for the opposing team. If making tackles that far down the field is as good as some say, then we should have kept Chowder and Wheels....they were doing the exact same thing.

How many times must someone watch Kiko rush and hit the wrong hole? Not sure, he keeps doing it on a regular basis.

Now the big kicker, he can't do what he's paid to, cover, both NE TD's in the game we won we're both because of Kiko, one he blew coverage as he always does, the other he was just standing watching the play go by....as he always does, now if we could have made the NE patsies start at the 20, then Kiko might have made the tackle.

But obviously by reading all the post here on this thread, most see exactly why Kiko is so bad. Some have posters on their wall, others just simply watching and can see exactly how bad kiko is....

 

this is as good a place as any to remind everyone that his cap number next year is $10 million

 

Yep. This is why I don't like to get into trying to grade Tannehill, or any other QB...I don't know enough about coverages and routes. However, even without knowing those things, it's kind of common sense. You see a ball that is way away from a receiver, I generally wouldn't think it was the fault of the QB, unless it was like a Tim Tebow throwing it. Many people though, see a way off target throw, and automatically default to "bad throw" because they assume the QB missed a receiver, and don't think about where the receiver was supposed to be. If a receiver is three yards off his mark, and the QB missed the mark by 2 feet, now you've got a ball that looks 11 feet off the mark, and people rant about the QB, when in reality, the receiver was far more at fault.

 

Sounds like a job for you...

 

Good thing is in 2019 there's zero dead money with his release so at least this will be his last year here.

 

Looks like I didn't look far enough into it. My bad.