I think it's called 'trimmed mean,' though I think there are other names for it, or other methods that eliminate outlier datapoints in order to calculate the mean/average.
Truth
Hell, can Philly fans take away a couple of his short/lost-yardage runs to rationalize him doing better than he did?
Philly fans?
When you take away all his gadget runs, he’s averaging well less than 4.0ypc.
Not to mention the 1st down run rate with Bucky in, is significantly lower than it is for Raachad.
Interesting stats thus far, White has been hit behind the line of scrimmage on 76% of his carries.
Biggs, you might just want to leave this topic all toghther
Actually, you can. I think it's called 'trimmed mean,' though I think there are other names for it, or other methods that eliminate outlier datapoints in order to calculate the mean/average.
Can you show me where this is the standard policy in football? Median and mode aren't a thing in football, while mean is.
Hell, can Philly fans take away a couple of his short/lost-yardage runs to rationalize him doing better than he did? I'm guessing not.
Anyway, in this case, it would be acceptable to remove the 60 yd rush (and the worst rush, if necessary), for the purpose of trying to more accurately calculate the player's rush yds/carry, in context of this game's performance by Saquon, as it was the greatest outlier, and skews the data to the point of making it unusable for the purpose intended. And when we do so, we arrive at a trimmed mean result of about 1 yd/carry, which is of course much closer to what he actually did on the field, and we just note the outlier 60 yarder as an exception.
Is it acceptable to remove Baker Mayfield longest pass completion when trying to rationalize he doesn't push the ball down the field enough or dumps it off too much?
Is it acceptable to remove Chris Godwin's longest pass reception when trying to rationalize he doesn't go deep enough or break enough tackles?
These aren't arguments I'm making. I'm simply highlighting that it's kinda foolish to make arguments like it. Saquan Barkley's YPC is his YPC. Baker Mayfield's YPA is his YPA. Chris Godwin's YPR is his YPR.
Yeah, I'm fully aware that YPA/YPC/whatever are what they are. I don't think you're not understanding what I'm saying, or maybe I wasn't clear enough. The idea behind eliminating outliers is to make the dataset more useable for specific hypotheses, and in this case, we're trying to find the most accurate description of Saquon's rushing performance last week. If you keep that 60-yd rush, you get 8.4yds/carry, which not coincidentally, isn't a value that's close to what he actually did on the field per rush, right?
I mean, if we actually go into the play-by-play and note all of his rushes how many are at/near 8.4yds/rush? So, let's see: Saquon's rushes went something like, "0, 3, 1, 7, 3, 4, 5, -1, 1, 59, 3 . . ." Basically none, right? So how valuable is that "8.4" number to us? It's not at all, right?
Generally, it's acceptable based on what it is you're trying to solve for. You could take it the other way and look for "the average of all the player's longest rushes," for example. Then you'd just eliminate all the lowest rushes and the 'average' rushes and make your calculations with just the longest ones, right?
No one is saying you do it in every instance. We aren't calculating the average the way we learned it in 5th grade. We're calculating for a specific purpose, when necessary. That's the mistake most people make when this is brought up. Trimming datapoints is a method for specific purposes. And most of the time, it's used for very small datasets; the impact of a few outliers in very large sets tend to cancel out, obviously.
We're calculating for a specific purpose, when necessary. That's the mistake most people make when this is brought up. Trimming datapoints is a method for specific purposes
Biggs understand that, I would think.
He's just caught himself and his partners in amess because they are SELECTIVELY trimming the dataset while simultaneously attacking anyone who disagrees.
That's called HYPOCRISY
To Biggs "and his ilk" trimming the datatset is "reasonable" when you want dont want to admit that White is likely to lose carrier to Irving. If that is what you are trying to refute, you don't compare YPC to YPC (the rule that BIGGS just posted lol).
I'm simply highlighting that it's kinda foolish to make arguments like it. Saquan Barkley's YPC is his YPC. Baker Mayfield's YPA is his YPA. Chris Godwin's YPR is his YPR.
You say, well take away Irving's "gadget plays" or take away White's "1st down runs." In that argument the "context" is importnat so trim the dataset.
But, if you want to refute that the Bucs largely contained Barkley, you FLIP on your own stated rule (LMAO) and advocate AGAINST trimming the data set, as Biggs is doping in this thread NOW.
Hilarious really because its not about statistics, it is about trolling. About arguing to argue, even if your posts are TRANSPARENTLY hypocritical.
Philly fans?
Yes, dipshit. Saquan Barkley plays for the Philadelphia Eagles. "Philly" is short Philadelphia. "Fans" is short for fanatics.
That will be all of my spoon-feeding the toddler for today.
I don't think you're not understanding what I'm saying
I understand exactly what you're saying. It's just not a thing that is done in the NFL, except when fan-bases are trying to rationalize their player/team performing better than they actually did. See below:
So how valuable is that "8.4" number to us? It's not at all, right?
Of course it is. It tells us if Barkley was given an 11th carry (which he should have been given more carries with those WR's out) he may break another 60 yarder and he'll then have a 13.1 YPC. Or he may not. That's how averages work and why you don't see, and will never see, the nonsense your talking here on any serious football website. The players get what they get.
So, let's see: Saquon's rushes went something like, "0, 3, 1, 7, 3, 4, 5, -1, 1, 59, 3 . . ." Basically none, right? So how valuable is that "8.4" number to us? It's not at all, right?
unless . . . .
Bucs didn't keep Barkley in check. He had 116 yards on 12 touches...
FFPTs - 33.2, 17.6, 33.6. . . . 13.6
Point, his running was subpar (for him, this year) and his receiving about average. He is a special talent BUT:
Score: 33-16
They did keep him in check, as much as anyone can keep him in check
he had similar stats when they BARELY lost to the falcons so the absence of the other players certainly plays a big part, so certainly agree there BUT the Bucs implemented their game plan yesterday, imposed their will on the Eagles, so give credit where due.
He had 8.4 yards per carry...
60 came on one carry so obviously the average is going to be skewed.
That’s how averages work. You don’t get to pick and choose what plays to count.
We're calculating for a specific purpose, when necessary. That's the mistake most people make when this is brought up. Trimming datapoints is a method for specific purposes
Biggs understand that, I would think.
He's just caught himself and his partners in amess because they are SELECTIVELY trimming the dataset while simultaneously attacking anyone who disagrees.
That's called HYPOCRISY
To Biggs "and his ilk" trimming the datatset is "reasonable" when you want dont want to admit that White is likely to lose carrier to Irving. If that is what you are trying to refute, you don't compare YPC to YPC (the rule that BIGGS just posted lol).
I'm simply highlighting that it's kinda foolish to make arguments like it. Saquan Barkley's YPC is his YPC. Baker Mayfield's YPA is his YPA. Chris Godwin's YPR is his YPR.
You say, well take away Irving's "gadget plays" or take away White's "1st down runs." In that argument the "context" is importnat so trim the dataset.
But, if you want to refute that the Bucs largely contained Barkley, you FLIP on your own stated rule (LMAO) and advocate AGAINST trimming the data set, as Biggs is doping in this thread NOW.
Hilarious really because its not about statistics, it is about trolling. About arguing to argue, even if your posts are TRANSPARENTLY hypocritical.
Ah, okay, makes more sense to me now.
Though to be fair, Las Vegas does stuff like that all the time. They have to, otherwise some of their lines would be massively skewed, particularly early in a season, when there just isn't much data to begin with. But obviously if it's done with the wrong intent, you get skewed results anyway, just in a predetermined direction, which is what it sounds like some people are doing to troll.
With that, I'll gracefully bow out of this part of the discussion. Thx for the insight.
Philly fans?
Yes, dipshit. Saquan Barkley plays for the Philadelphia Eagles. "Philly" is short Philadelphia. "Fans" is short for fanatics.
That will be all of my spoon-feeding the toddler for today.
I don't think you're not understanding what I'm saying
I understand exactly what you're saying. It's just not a thing that is done in the NFL, except when fan-bases are trying to rationalize their player/team performing better than they actually did. See below:
So how valuable is that "8.4" number to us? It's not at all, right?
Of course it is. It tells us if Barkley was given an 11th carry (which he should have been given more carries with those WR's out) he may break another 60 yarder and he'll then have a 13.1 YPC. Or he may not. That's how averages work and why you don't see, and will never see, the nonsense your talking here on any serious football website. The players get what they get.
Uh, averages/means don't do "ifs," for one thing. So whether or not he got an 11th carry is a moo point. The thing we're trying to solve for is what his actual yds/carry average was in that particular game. And when we set aside the 59 yarder, we arrive at 2.7 yds/carry, which falls almost dead center of what his other 9 carries went for, right? So which is the more useful number?
Anyway, just trying to help. Now I'll bow out of this part of the conversation.
Uh, averages/means don't do "ifs," for one thing
Then why are you people making the "IF you take away the 59 yarder from Saquan" foolish argument?
Nevermind, I know why.
The thing we're trying to solve for is what his actual yds/carry average was in that particular game.
It's not difficult to solve. His actual YPC for that particular game was 8.4 YPC. This isn't an opinion.
Uh, averages/means don't do "ifs," for one thing
Then why are you people making the "IF you take away the 59 yarder from Saquan" foolish argument?
Nevermind, I know why.
The thing we're trying to solve for is what his actual yds/carry average was in that particular game.
It's not difficult to solve. His actual YPC for that particular game was 8.4 YPC. This isn't an opinion.
The 59 yarder was an actual result, just an outlier. What you're proposing is a hypothetical result, one that never happened. That's the difference.
Uh, averages/means don't do "ifs," for one thing
Then why are you people making the "IF you take away the 59 yarder from Saquan" foolish argument?
Nevermind, I know why.
The thing we're trying to solve for is what his actual yds/carry average was in that particular game.
It's not difficult to solve. His actual YPC for that particular game was 8.4 YPC. This isn't an opinion.
The 59 yarder was an actual result, just an outlier. What you're proposing is a hypothetical result, one that never happened. That's the difference.
You're proposing a hypothetical result as well, because you're attempting to take away a result that did actually happen. That's the whole point.
Let's take away Hurts' sacks and fumbles. Those were outliers. The defense didn't really perform as well as everybody thinks when you remove those outliers.
Uh, averages/means don't do "ifs," for one thing
Then why are you people making the "IF you take away the 59 yarder from Saquan" foolish argument?
Nevermind, I know why.
The thing we're trying to solve for is what his actual yds/carry average was in that particular game.
It's not difficult to solve. His actual YPC for that particular game was 8.4 YPC. This isn't an opinion.
The 59 yarder was an actual result, just an outlier. What you're proposing is a hypothetical result, one that never happened. That's the difference.
You're proposing a hypothetical result as well, because you're attempting to take away a result that did actually happen. That's the whole point.
Let's take away Hurts' sacks and fumbles. Those were outliers. The defense didn't really perform as well as everybody thinks when you remove those outliers.
Uh, I don't think your use of the word 'hypothetical' is accurate. Or, you're misunderstanding my usage, either way.
Uh, I don't think your use of the word 'hypothetical' is accurate.
Please, explain "hypothetical" to me. Which of these is the hypothetical outcome and which of these is the actual outcome of Saquan Barkley's week 4 performance?
Option A - 10 carries for 84 yards
Option B - 9 carries for 25 yards
That was easy one, but this one may be more difficult for you. Are you hypothetically bowing out of this discussion or actually bowing out of this discussion? It's hard to tell, because you've now said that you are bowing out twice - yet you can't stop yourself from posting.
That was easy one, but this one may be more difficult for you. Are you hypothetically bowing out of this discussion or actually bowing out of this discussion? It's hard to tell, because you've now said that you are bowing out twice - yet you can't stop yourself from postin
there it is
