Stat Glitch

[Bonegavel]

Nate didn't play much in 2005, but I guess when they told him to give 110%, he said, "No, no" and gave 114.2%.

http://www.bonegavel.com/images/stuff/natewayne.gif

I highlighted the year in question (for those not familiar with the game)

[Chubby]

impressive, 7 plays he was in on and 8 tackles :)

[Easy Mac]

was he on special teams?

[Bonegavel]

He could have been (his special team rating is ~45), but that shouldn't show up in here?

[Easy Mac]

It would if he made a tackle, no? Offensive players get tackles if they tackle someone who gets a fumble/INT.

[sabotai]

Bone, I think the game adds special teams tackles to the tackles stat.

[Bonegavel]

So it is dividing all tackles by only defensive plays played? That is odd.

[sabotai]

Yeah, it would seem like that. An oddity that Jim must not have been aware of. Maybe he should adjust it somehow.

Don't matter to me though. I never paid attention to the tackle percentage thing anyway.

[Bonegavel]

I use it as a tie-breaker (in the case of releasing or signing) for players that are very similar. With 114.2%, I should give him a raise.

[Bonegavel]

Well, I'm glad I saw this and now know that the TkPct should be ignored.

[edit: I'm going log-file diving for a minute to see if I can find out what is going on here.]

[finkenst]

tackle, forced fumble.
tackle the recoverer.

2 tackles, one play.

[sabotai]

I wouldn't say you have to ignore it. Just be aware of who your special teams players are, and ignore it for people who have only played a dozen or so plays.

[Bonegavel]

Found the offending log file:


Defense Tackles Assists RunPlay PassPlay
32 Lewis 10 2 27 28
51 Emmons 7 3 27 28
23 Vincent 7 0 27 28
90 Simon 6 2 27 28
98 Spencer 5 2 20 21
26 Sheppard 4 2 27 28
92 McDougle 4 1 27 28
20 Dawkins 2 0 27 28
65 Green 2 0 20 24
79 Yoo 1 1 18 13
55 Harrison 1 0 6 4
24 Brown 1 0 1 3
54 Wayne 1 0 0 0
52 Gardner 0 1 22 17
38 Barber 0 1 5 11
91 Myers 0 0 12 14
96 Berry 0 0 4 5

[Bonegavel]

and, in the log file, here it is:


PHI 2 Akers kicked off 65 yards from the PHI30.
KCY 41 Myles returned the ball 14 yards to the KCY19. Tackled by PHI 54 Wayne.

[Bonegavel]

Looks like Jim needs to add a few stats to to the Defense stat page: KPlay to track the kick off plays, and keep a separate tackle total for them. I think it is safe to say that tackles on Kickoffs are different enough from a regular Def play to want to see them separate.

[oykib]

Also, I'd like to see how my kicker is doing on kickoffs without having to eyeball it in gamelogs.

[QuikSand]

Originally posted by Bonegavel
Well, I'm glad I saw this and now know that the TkPct should be ignored.

I don't know - ignoring seems a little harsh. Since its value is mostly relative anyway (what does a 13.3 mean on its own, except when placed beside a 12.2 and a 15.8?) I guess the only thing is to recall that guys on ST (especially gunners) will probably get a few extra tackles, and have this number inflated. (Note - the inflation occurs in the raw tackle total also - it's not just a function of the TkPct calculation)

Doesn't seem too big a deal to me -- I'll still use it to give myself a very general sense of how much a part of the defense a player ends up being.

[GoldenEagle]

Thats why he was a starter next year. :)

[Bonegavel]

As always, quick is the voice of reason. Good points, all. It would be nice to see it separated, but at least now I know there may be a slight a discrepancy in the TkPct number and actual defensive play.

[Ben E Lou]

I've noticed this before, and it isn't a huge deal to me either. As someone said earlier in the thread, TkPct is usually more of a "tie-breaker" for me in determining who to start/etc. Even if special teams wasn't an issue, it certainly isn't a statistic that can be used with any level of confidence on a player who was in on less than 10 defensive snaps for the season. With that small number of plays, it is much more possible that the offense ran toward his side 80% of the time, or something like that.

[Bonegavel]

Now, I also figure that in an average game they might participate in 3-6 kickoffs, so a player that sees significant play won't be thrown off too far with an extra 6 tackles added to his total without a matching RPlay or PPlay. (it was just the 114.2% that made my drew my attention)

That was me regarding the tie-breaker. I see the TkPct being most meaningful for LBs, because, LBs are supposed to be guys making the tackles. Hopefully, I assume that a high TkPct (relatively speakin) means that the LB was doing his job and it only really means something if the point difference is egregious.

[QuikSand]

I tend to watch TkPct for defensive tackles, too -- I like to see a starting DT get 40 or more tackles and TkPct of around 8 or so. For that position, it's tough to see how a tackle can be a bad thing. Sometimes, with excellent DTs against the run (especially just one of them) I'll see one get 60-70 tackles and a TkPct of over 10... stellar stuff.

[Sidhe]

Well I hope you did notice that whenever he was on the field he made the tackle. I would see how he performs as a starter..

This might be a really big hint that you have a breakout player on your hands.

[Ben E Lou]

Originally posted by Sidhe
Well I hope you did notice that whenever he was on the field he made the tackle.Not necessarily true. BG showed us a box score of one game in which he had a special teams tackle but never played on defense. It is fully possible that some/all of the other tackles he made were in games in which he played a few snaps on defense, but the actual tackles were made on special teams. In other words, just because he played one defensive snap in a game and recorded one tackle doesn't mean that the tackle was made on that defensive snap.

[QuikSand]

Originally posted by Sidhe
Well I hope you did notice that whenever he was on the field he made the tackle. I would see how he performs as a starter..


I think you are missing the point of the thread.

He did not make tackles on every play he was involved in. In all likelihood, he made nearly all of those tackles on special teams --like the one indicated above. During the 7 plays he actually went onto the field - we may ave made 1 or 2 tackles (or none) -- but we certainly cannot say that he made 7 tackles. That's the "glitch" mentioned in the thread title -- this gives a mistaken impression, which becoms very misleading when a player is in for very few defensive plays but does play on special teams.

[cthomer5000]

Originally posted by QuikSand
I tend to watch TkPct for defensive tackles, too -- I like to see a starting DT get 40 or more tackles and TkPct of around 8 or so. For that position, it's tough to see how a tackle can be a bad thing. Sometimes, with excellent DTs against the run (especially just one of them) I'll see one get 60-70 tackles and a TkPct of over 10... stellar stuff.

Same here. I play a 3-4, so I need a nose tackle to just clog up the middle and get in the way of as many things as possible. In their best seasons, they'll get between 50-70 tackles, 8-12 sacks, and a pretty respectable number of hurries and assists. It makes life easier for all around them (including my generally shitty DE's).

[fantastic flying froggies]

Special team plays are not tracked at all... They 're lumped in with all other plays...