2021 RCF Mock Draft: Discussion, Trades and Miscellany

[Jack Brickman]

You have to provide a reason why.

The reason has been provided a dozen times over in this thread. A site did research on years' worth of drafts where prospects ran at both the Combine and their pro day and found that the average difference was .05 seconds. And so they add that time to pro day times as a consequence of that research.

I think we can all agree that this system isn't completely perfect, but I'd argue it makes more sense than not adjusting the times at all when we know they're probably inaccurate to some degree. Not every time would be exactly .05 seconds faster. Some might be .03. Some might be .08. It is, again, an average.

What the CBS draft profiles have done is just allow a .05 variance for all NFL prospects this year. So, for example, they say Micheal Carter is between a 4.50 and 4.55.

I think this is the most responsible way to do it, because just adding .05 doesn't solve the human error.

And that's a fine system if you're just listing times. However, if you're running a database, you can't input 4.50 to 4.55 into the field for forty time. You need an actual number, not a range, and so it was decided to add the average variation between forty and pro day times to each prospect who opts to not run at the Combine (or there is no Combine like this year).

 

[Out of the Rafters at the Q]

The reason has been provided a dozen times over in this thread. A site did research on years' worth of drafts where prospects ran at both the Combine and their pro day and found that the average difference was .05 seconds.

So, in order for this adjustment to be justified, the next step is to prove that UNC's pro day was "average."

Without any outside data points to use for comparison, I am not sure how one could make this conclusion. Therefore, the responsible thing is to not make the adjustment and be clear in your methods.

 

[Randolphkeys]

I just think it's disingenuous to see a 4.50, and then claim it's really a 4.59 without really having anything to back that up.

Exactly. Any database that is that far off just in the name of " we need an actual number" isn't very useful in my book.

 

[Out of the Rafters at the Q]

Exactly. Any database that is that far off just in the name of " we need an actual number" isn't very useful on my book.

Honestly, I could care less about a 40-time. I'm more interested in the 10 and 20-yard splits.

But when they're inaccurate in one metric, it makes me have less faith in their other metrics.

You tell me a guy ran 4.50 at his pro day, and I understand what that means. You list someone's 40-time as 4.59, and that is a completely different story.

 

[Jack Brickman]

So, in order for this adjustment to be justified, the next step is to prove that UNC's pro day was "average."

Without any outside data points to use for comparison, I am not sure how one could make this conclusion. Therefore, the responsible thing is to not make the adjustment and be clear in your methods.

The UNC pro day was, like all pro days, hand timed. That's the issue here. Timing by hand is inevitably going to be less accurate than the system the Combine uses because movement is imperceptible to the human eye for a fraction of a second.

This would generally be irrelevant. Unless, of course, you're running a database full of official Combine times and trying to compare prospects in historical context. At that point, you need to do the responsible thing and adjust the times of those who didn't run at the Combine accordingly.

 

[Out of the Rafters at the Q]

The UNC pro day was, like all pro days, hand timed. That's the issue here. Timing by hand is inevitably going to be less accurate than the system the Combine uses because movement is imperceptible to the human eye for a fraction of a second.

This would generally be irrelevant. Unless, of course, you're running a database full of official Combine times and trying to compare prospects in historical context. At that point, you need to do the responsible thing and adjust the times of those who didn't run at the Combine accordingly.

Guess everyone at the UNC pro day just had really slow hands at the beginning of the 40, but really jumpy trigger fingers at the end. I bet your guy knew all this, which is why he decided to arbitrarily add .09 seconds to a 40-time and report it as accurate.

 

[Jack Brickman]

Guess everyone at the UNC pro day just had really slow hands at the beginning of the 40, but really jumpy trigger fingers at the end. I bet your guy knew all this, which is why he decided to arbitrarily add .09 seconds to a 40-time and report it as accurate.

First, it's not "my guy". This is a professional website and one of the best and most respected fantasy football resources available. This is a business that employs data scientists, not some asshole in a basement throwing numbers at a wall. This is a site that has been at the forefront of the analytical movement, and was an early adopter of statistics like breakout age and dominator rating. I believe they actually created the concept of teammate score, which attempts to numerically quantify the quality of a player's teammates in college to help adjust for lower dominator ratings as a result (Terrace Marshall has the highest teammate score they've ever logged, for example).

Second, they have Carter listed as a 4.54 pro day time adjusted to a 4.59. It literally shows both numbers on the site. Nothing is being hidden and the site makes it clear that one time was the time measured at his pro day.

 

[Out of the Rafters at the Q]

Sure seems like, if you're going to report something other than 4.50 (like 4.54 or 4.59) you should have justification. Here's just the top google results that I clicked through.

https://goheels.com/news/2021/3/29/carolina-football-pro-day-results.aspx 40-yard: 4.50
https://www.tarheelblog.com/2021/3/...l-carter-javonte-williams-dyami-brown-newsome 40-yard Dash: 4.50 s
https://247sports.com/college/north...z-Surratt-Dazz-Newsome-Dyami-Brown-163300783/
40-yard dash - 4.50 (10-yard split - 1.63)
https://247sports.com/college/north...mi-Brown-Dazz-Newsome-40-yard-dash-163283505/
40-yard dash - 4.50 (10-yard split - 1.63
https://walterfootball.com/proday.php 4.50
https://dknation.draftkings.com/nfl...-day-news-2021-running-backs-fantasy-football 40-yard dash: 4.50
https://www.cbssports.com/nfl/draft...er-with-measurements-times-for-top-prospects/ He also posted a 4.50 second 40-yard dash
https://www.si.com/nfl/draft/player...l, posting a,important than all the accolades. Carter ran well, posting a 4.50 40-yard dash and a 3.98 20-yard shuttle
https://www.pff.com/news/draft-2021-nfl-draft-comps-north-carolina-rb-michael-carter He clocked in at 4.50 seconds on the 40-yard dash

I'm sure there's a reason he input 4.54. I'm sure there's justification for why 4.59 is the best combination of accuracy and simplicity for him to use in his database. But, if we're simply talking about a specific player who ran 4.50 at his pro day, saying he ran 4.59 is just not correct or helpful.

 

[adam81king]

Yall are *still* talking about the variations in measurements???

The real draft can't come soon enough...

 

[Jack Brickman]

Sure seems like, if you're going to report something other than 4.50 (like 4.54 or 4.59) you should have justification. Here's just the top google results that I clicked through.

https://goheels.com/news/2021/3/29/carolina-football-pro-day-results.aspx 40-yard: 4.50
https://www.tarheelblog.com/2021/3/...l-carter-javonte-williams-dyami-brown-newsome 40-yard Dash: 4.50 s
https://247sports.com/college/north...z-Surratt-Dazz-Newsome-Dyami-Brown-163300783/
40-yard dash - 4.50 (10-yard split - 1.63)
https://247sports.com/college/north...mi-Brown-Dazz-Newsome-40-yard-dash-163283505/
40-yard dash - 4.50 (10-yard split - 1.63
https://walterfootball.com/proday.php 4.50
https://dknation.draftkings.com/nfl...-day-news-2021-running-backs-fantasy-football 40-yard dash: 4.50
https://www.cbssports.com/nfl/draft...er-with-measurements-times-for-top-prospects/ He also posted a 4.50 second 40-yard dash
https://www.si.com/nfl/draft/player-interviews/interview-with-north-carolina-running-back-michael-carter#:~:text=Carter ran well, posting a,important than all the accolades. Carter ran well, posting a 4.50 40-yard dash and a 3.98 20-yard shuttle
https://www.pff.com/news/draft-2021-nfl-draft-comps-north-carolina-rb-michael-carter He clocked in at 4.50 seconds on the 40-yard dash

I'm sure there's a reason he input 4.54. I'm sure there's justification for why 4.59 is the best combination of accuracy and simplicity for him to use in his database. But, if we're simply talking about a specific player who ran 4.50 at his pro day, saying he ran 4.59 is just not correct or helpful.

There are numerous sites that list Carter as running a 4.54 as well, so I'm not sure what your point is. When multiple people time a guy by hand, you're invariably going to get different results.

 

[Out of the Rafters at the Q]

There are numerous sites that list Carter as running a 4.54 as well, so I'm not sure what your point is. When multiple people time a guy by hand, you're invariably going to get different results.

I'm sure there are. Like I said, I just Google'd "Michael Carter pro day" I think... may have been "Michael Carter 40 time" or something like that and clicked down every link on the front page. I pasted the URL here, and copied the text about his 40. As you can see, they all said 4.50.

I'm sure the other sites have reasons for 4.54. I'm sure your guy has valid reasons for inputting 4.59 into his database. I'm not questioning those.

To say Michael Carter ran a 4.59 is disingenuous. Like, even if your guy's reasons are valid, he's not even claiming that Michael Carter ran a 4.59. He's claiming that, for the sake of historical comparison, 4.59 is the figure he's inputting into his database.

Did Etienne and Gainwell go back to school?

Carter just seems like a plodder to me. He looks slow on tape and ran a 4.59.

That NC pro day was a disaster, by the way. Both Carter and Williams were slower than you'd want to see and Williams came in 15 to 20 pounds lighter than expected.

This just isn't true, and is why we're three pages down this rabbit hole.

 

[Surge]

How big are his thighs?

 

[adam81king]

I'm setting the over/under at 29.5 for number of posts talking about a site adding half a second to 40 times. Without counting, anyone wanna make some bets?

 

[Out of the Rafters at the Q]

I'm setting the over/under at 29.5 for number of posts talking about a site adding half a second to 40 times. Without counting, anyone wanna make some bets?

Over.

I know it's at least three pages.

 

[sportscoach]

How big are his thighs?

Mind are bigger!! @The Oi

I'm setting the over/under at 29.5 for number of posts talking about a site adding half a second to 40 times. Without counting, anyone wanna make some bets?

At the end of the day, the argument is irrelevant in my mind. Who cares if it's 4.54 or 4.59, how does his game film look? That's what matters the most!