O-Line Blocking Percentages
 

When you are evaluating O-linemen, how much consideration do you give to a player's blocking percentage stats?
Blocking percentages seem to vary quite a bit from year to year for individual players, even players with nice long bars for run blocking. The stats almost seem to be too unreliable to use as a major factor in evaluating free-agents. Am I missing something?

0% stats.

100% red bars.

That's how every single player with ~3+ years of experience should be evaluated. Always. There are no hidden "ZOMG HE'S REALLY GOOD EVEN THOUGH HE LOOKS BAD" or "ZOMG HE'S REALLY BAD EVEN THOUGH HE LOOKS GOOD" things to worry about in FOF once players are at or near red=green. If he's got good ratings and isn't playing well over a long period of time, it's one or more other things that you're doing poorly. (Game plan, chemistry, cohesion, staff, talent around him, etc.) If he's got bad ratings and he is playing well over a long period of time, it's one or more other things that you're doing well. (Game plan, chemistry, cohesion, staff, talent around him, etc.)

Dola:

When it comes to that particular OL stat, a huge factor is the dude totin' the rock. (See: talent around him.)

I have been call an o-linemen snob because my standards for run blocking are high. I prefer an individual block % between 40% and 45% (or higher)!!!

In the IFL, my team KRB % was 42.9% and 47.0% over the last two seasons. And that means the individual KRB % are high for most of my linemen. Of course, you do need big red bars, some chemistry and the right RB(s) to make that happen. The run block bar average for the 5 lineman is over 87.

This is a smash mouth team, BTW.

Somewhere buried on this board is a thread that uncovered what is still 100% true... in FOF, successful run blocks are essentially a play stat, not a player stat.

It works this way:

-on every called run play, a "key blocker" is selected (that's the key run block opportunity) ... and we presume that blocker has some heightened role in the play's outcome but we can't prove/refute that
.....(the selection of the lead blocker may or may not match up to who you think it would be)

-after the run play is resolved, the game determines by a "win/fail" system at least similar to something developed by Football Outsiders whether the play was successful (based on down a distance the play must accomplish a certain % of the yards-to-go)

-if the play was successful, the player pre-selected for the key run block opportunity will be awarded a "key run block"

So, if you buy the above, then KRB% is really just "what share of run plays behind this guy were successful?" And that's a lot different from "how often did this guy do a great job?" If a star RB makes a lot of guys miss, his lead blockers are going to ex post facto be credited with lots of KRBs because the RB was good... and the contra as well.


So... the winner and STILL champion... big red bars.

all they said above -- and if you want to evaluate player performance as you asked above, the best way is to look at the +++ and --- ratings in the detailed game logs. but they also depend on usage, game plan, etc

I have always assumed Quik's description was accurate, including the team aspect of the run outcome (both o-line as a unit and the RB). Jim also wrote the "hints" below in the help - note the reference to the "team's performance" and the run direction weighing.

For defensive ratings against the run, I compiled a grid of responsibility for each defensive position. For instance, the left defensive end had 25 percent of his run rating based on his team's performance against runs around right end and only 4 percent of his rating based on runs around left end.

Thanks everybody.
That'll save a lot of time.

Can we make the same assumption about Dback stats like PDpct? I noticed that those also don't seem to square up with bar length in an awful lot of cases.

PDPct is something we also figured out a while back... I don't really recall the details.

IIRC, it's a formula that starts at 82.5, and then grants a small addition for good stat outcomes (interceptions and passes defensed) and a small subtraction for bad stat outcomes (completions allowed) all on a per-pass-play basis.

That is why a lot of defensive linemen, rarely in coverage, will have ratings very close to 82.5 (which can make them look "good" when your typical starting DB might have a rating of 76 or so). They have very few of any of those events to move their number far from the origin.

So, PDPct is just an amalgamated number of stats that we already see and nominally understand.

(now, the related question would be how much do the coverage ratings for the defensive player tagged as "allowing the reception" matter in the dice-rolling that determines whether the play is successful? that's the black box, but most assume there's a "coverage cloud" concept at work, where the overall effectiveness of your players on the field contribute to that outcome, not just the one player.)

I'd suggest reading this thread: How do you all evaluate defensive backs in FOF8? - Front Office Football Central

Super helpful. I'm assuming this logic applies to ALL players, not just O-line?

I think that is a safe and wise assumption.

Pd% equation is (825 - ((x*pass defended)+(y*int)-(z*catches)) / pass plays) / 10.

I have the exact formula on my PC, I'll edit this post with it later.

The person who gets the stat on D whether positive or negative is based on a distribution that is why your SLB may defend a flat pass to the X or similar.

Same with run stats, in all cases however the distribution is not entirely random. You can look at the run defense text mentioned from the help file to get a rough idea.