Virginia Tech leads the nation in winning % forfeited

This is good, right? Not bad for a staff in Year 6 Year 1, right?

DISCLAIMER: Forum topics may not have been written or edited by The Key Play staff.

Comments

We're number 1! We're number 1! We're number 1!

Mission Completed!

We put the K in Kwality

Here we go, the final play of the game. IT’S INTERCEPTED! IT’S INTERCEPTED BY BANKS! HE’S RUNNING IT ALL THE WAY DOWN THE SIDELINE! THE GAMECLOCK WINDS OUT! ANTONIO BANKS TAKES IT IN!
https://youtu.be/LgEAr3XFrcI

Not great Bob!!!!!!! Looks like "expected outcomes" has its downside. Too conservative.

Minor nitpick: I don't think the graphs make any distinction between kicking on a "goal to go" down early or late in the game. Thread title is kinda misleading because of that?? Obviously win probability is affected more later in the game when leverage is higher, but not sure this is "late game" anything in a vacuum unless I'm missing something?

"Why gobble gobble chumps asks such good questions, I will never know." - TheFifthFuller

Yeah this actually might be over the course of the game and not just late. I saw the part about the timeout being in the 4th Quarter and assumed it was for late game.

Regardless, you're dead-ass on. For being a guy who stresses "expected outcomes" to literally be leading the nation in this stat because of pure coaching decisions is... Well, it's bad.

I will not donate to Virginia Tech Athletics as long as Justin Fuente is the head coach of VT Football. Enough is enough.

Looks like "expected outcomes" has its downside. Too conservative.

It doesn't have a downside, Fuente just doesn't know how to calculate expected outcomes.

I really like a lot of what Fuente says - hard smart tough, Grit, expected outcomes - are really smart, level headed principles for any organization/individual. The problem is that he doesn't actually act on any of this stuff.

Twitter me

The problem is that he doesn't actually act on any of this stuff.

Perhaps one might say that he doesn't actually execute on any of this stuff ;)

đŸ”„â›Č

If you feel the leather in your hand let it rip.

“Also, a microwave has never danced it's ass off to Jackie Wilson.” - AssPocketFullOWhiskey

Exactly. It all sounds great, and I'm sure Fuente is a great guy. Problem is none of it is working out. And being a nice guy shouldn't be the controlling factor in a multi-million dollar activity.

After listening to the TSL podcast today and reading a non-paywall thread over on 247 discussing the leadership/admin views of Fuente, I'm bracing myself now for no change or a very underwhelming one if we even do. Starting to think the rot starts at the very top. Whit very well may have been overruled last year. Apparently being a nice guy is the top criterion for being a P5 football coach. I'm in the wrong line of work.

If Fuente isn't fired after this season it's going to lead to lasting damage to the football program and, by extension, the whole athletic department. If leadership doesn't understand this they'll just learn the hard way.

đŸ”„â›Č

Brace yourself. It appears that many in the higher administration really like Fuente and think the incoming cash from the Reach campaign is the springboard he needs to succeed. Could be totally off base, but that's what it sounded like.

I really hope it's off-base. Fuente might be likeable but no amount of cash is going to fix the problems VT's football program has with him at the helm. Cash isn't going to fix a regressing offense year over year. Cash isn't going to fix consistently back-breaking in game decision making. Cash isn't going to fix utter and complete lack of development of QBs, WRs, and RBs. Cash isn't going to fix incredible volatility in performance from game to game. Cash isn't going to fix QBs broken by the nature of our offensive play calling and strategy. Cash isn't going to fix the revolving door that is the transfer portal. Cash isn't going to fix consistent execution errors that cost us games. Cash isn't going to make our offense better. Cash isn't going to make our team better. Cash is going to line the pockets of coaches who aren't earning their keep. That's about it. Reach for excellence is destined to crash and burn if Fuente is retained.

đŸ”„â›Č

Agree. If that's truly the sentiment about Fu from the people in charge, it's going to take a less than half capacity Lane and donations really drying up to change things. It sounded as if the chief concern with football from the administration was just running a clean program and don't embarrass the university. Including some major revisionist history about Beamer. I think Frank is a great guy and high character, but let's not pretend that he never had a few bad apples on the football team or gave second chances. I remember the beginning of the tv broadcast in the '96 Orange Bowl talking about how VT and Nebraska were neck and neck for most arrests in the country. Don't give me this revisionist garbage that we never took a few risks with recruiting players.

Really smacks of laziness, cheapness, and either not knowing or wanting to know how to be a successful college football program.

It sounded as if the chief concern with football from the administration was just running a clean program and don't embarrass the university. Smacks of laziness, cheapness, and either not knowing or wanting to know how to be a successful college football program.

Running a clean program is admirable. It is also possible to run a clean program and be successful. It is also possible to fire coaches who are underperforming and still run a clean program. I agree that it sounds lazy, cheap and naĂŻve to bury one's head in the sand and demand that we run a clean program at all costs. This is Capitalist America where money is King. Money always wins at the end of the day. And, to your point, the money is going to dry up real quick if we continue on our current trajectory. It's not worth checking off a single box and calling it a day when that single box is "run a clean program".

đŸ”„â›Č

This would imply that those in power at VT think that running a nationally relevant winning football program and running a clean football program are mutually exclusive.

I certainly hope that's not the case, because it opens a whole big can of worms in assuming that players with elite levels of athletic skill are incapable of staying out of trouble, with some troubling undertones that I would really hope the leaders of our university don't believe.

I will not donate to Virginia Tech Athletics as long as Justin Fuente is the head coach of VT Football. Enough is enough.

I think that the administration just doesn't realize what it takes to win at a high level in P5. They think because Fuente runs a clean program and doesn't bring embarrassment to the university that we are doing things the right way, regardless of wins and losses. It seems they think he just needs more money and his vision will fall into place. That's completely overlooking the increasing evidence that he and his staff are just not good coaches. If VT were to come out and explicitly say, hey we're not going to get involved in the SEC-style recruiting and win at all costs mentality, we want to run a clean program and coach at a high caliber and compete in the ACC, I would be ok with that. Problem is, I don't think Fu fits the bill because he isn't a great coach. If we go that route, go hire Dave Clawson right now, because he's a hell of a lot better coach and would fit the mold of an academics-first university, except on a larger scale with more to offer.

I agree overall that more funding and resources are needed for the program to succeed. But I don't agree that those things alone will fix the cluster we are seeing on the field right now. If VT wants to align itself with an academics-first model but fund the program to compete on a top 25 level, that's good and fine. But Fu ain't it. Being a nice guy does not mean he's succeeding at being a good coach, recruiter, or developer of talent.

Fuente shared a candid glimpse into his in-game thinking and motivation. He manages games far too off by feel, emotion, routine, and other irrational thought processes, and not enough because of logic and fact-based rationale. This bothers me more than anything, coaching-wise, about him. He makes Mike London look competent over four quarters of football.

He coaches the players to err on the side of caution to emphasize expected outcomes as much as possible, going so far as to bench players for mistakes, but openly admits he shoots from the hip far too much with his own decision making in crucial situations?

That's.... Not great

I will not donate to Virginia Tech Athletics as long as Justin Fuente is the head coach of VT Football. Enough is enough.

where did he openly admit to doing this "far too much"?

He says this but really the only logical motivation is bad math/outsmarting yourself. The difference between going up by 1 or 2 is only meaningless if there's less than 5 min in the 4th. Any other time in the game you want that point. He shouldn't have been Going for two even before the penalty.

2 would have put us up by 3. At the time. Which is stupid itself, but then to go up by 7 after burmiester makes than run and not go for 2? WTF!!!!! If you have a chance to go up 9 then you do it.

Agreed, I liked the decision to go for two until the false start. After the false start, you kick the extra point.

I'm fine with him being aggressive and going for 2 there, not fine with him playing super conservative after. Be consistent either way

Going for 2 in the 3rd with your backup QB from the 7 is beyond dumb.

I can't believe this is real. What a f-ing douche. Explains a lot.

"The Big Ten is always using excuses to cancel games with us. First Wisconsin. Then Wisconsin. After that, Wisconsin. The subsequent cancellation with Wisconsin comes to mind too. Now Penn State. What's next? Wisconsin?" -HorseOnATreadmill

I feel like we always go in kicking situations tho...

Shouldn't have gone for it regardless of the penalty. Poor decision period

Seriously! When the next drive isn't the last of the game it makes zero sense.

Same reason you don't go for it on 4th and short in your own territory. It's low percentage and failing the puts you in a way worse situation than just punting/taking the point

I don't think this stat captures that play?

"Why gobble gobble chumps asks such good questions, I will never know." - TheFifthFuller

Far be it from me to defend Fu - and I am not trying to do so in a broad sense - but I can understand some of this given our total inability to pick up small amounts of yardage on 4th and short if someone other than Evans or Herbert is there to carry the ball. Cannot blame him for being conservative and kicking. Makes some sense.

Now as for why we cannot pick up 1 yard on any consistent basis - and the bleeping @#!%$ we line up in the Fu-ing shotgun to hand the ball off several yards from the line when to need 8 _ _ inches for a first down, I can 100% blame him for.

Sorry the shotgun snap on 3rd or 4th and short or goal-to-go is a massive pet peeve of mine. I was stunned when we actually ran the QB sneak. And, lo and behold, it worked! pretty much guarantees we wont see it again with Corn calling plays.

Recovering scientist working in business consulting

A lot of this is on Vice (and losing Nester/Hudson). The run block is atrocious and has been since the Middle Tennessee game.

Lining up in the shotgun is annoying but idk if we'd even pick up a lot of these first downs under center

So weird, Herbert feels like a lifetime ago

Lol, I think most of us can agree that Fuente probably needs to go, but some of y'all are so obsessed with the dude that he's just living in your head. This is a meaningless statistic.

I mean, what the hell does "ESPN FPI winning percentage lost by kicking in go situations through Week 6" even mean? And do you honestly think that that tells you anything about the relative competencies of coaching staffs? If so, #2 6-0 Iowa is in the top quarter of the graph along with us, while Kansas, the worst team in the country, is in the top ten in the nation - what does that tell you about the coaching staffs? *Spoiler: NOTHING, because this stat is meaningless.*

This is a meaningless statistic.

When you've got goal-to-go in a given situation, does kicking help you or hurt you? This measures that. When attempt the fg in these situations, it's likely to hurt us

"Why gobble gobble chumps asks such good questions, I will never know." - TheFifthFuller

but some of y'all are so obsessed with the dude that he's just living in your head.

Yes. This. Thank you.

"... I think he played his nuts off. And you can quote me on that shit."

Haha, that's fair, I admittedly am not a big advanced analytics guy. As a general rule of thumb though, I tend to not take a lot of stock in stats that rank Kansas higher than all of the top 5 teams lol.

This is a meaningless stat, but supports the conclusion I already made

if only there were a stat that would measure whether this stat is more or less meaningless than the phrase "he's living in your head"

Fu doesn't win enough to live in anyone's head.

21st century QBs Undefeated vs UVA:
MV7, MV5, LT3, Braxton Burmeister, Ryan Willis, Josh Jackson, Jerod Evans, Michael Brewer, Tyrod Taylor, Sean Glennon, and Grant Noel. That's right, UVA. You couldn't beat Grant Noel.

Fuente just has a knack for always making the wrong decision. I have defended him in the past as I think he's a competent coach but it's like his feels are way off in late game decision making. He just needs to hire a feels guy.

He just needs to hire a feels guy

Jerry Sandusky?

Leg up, but still:

Recovering scientist working in business consulting

... and what will Fuente get for all of this? $10,000,000 U.S. Dollars! An amazing prize for sucking.

What other business rewards sucking with this amount of compensation?

#MakeTheMove . . . "Vick, dashing back . . . here he comes again . . . Electrifying . . . and have you ever seen anything like this?"

What other business rewards sucking with this amount of compensation?

Hmmm... Vacuum cleaner manufacturing? The 'adult film' industry? That's all I can think of.

Hot damn! Sign me up then. (For the vacuum cleaner manufacturing position, that is.)

#MakeTheMove . . . "Vick, dashing back . . . here he comes again . . . Electrifying . . . and have you ever seen anything like this?"

Pretty much every Fortune 500 company. Every top executive has a golden parachute for when it's time for the company to move on.

i can't find the most appropriate original clip for this, but Mel Brooks nailed this one - A flop will make more than a hit.

We're analyzing the hell outta this, but just looking at various teams along the graph and where they are makes this feel meaningless (as others have pointed out). Alabama is one of the worst, and up until this week was #1 and undefeated. NC State is ranked and 4-1 and also pretty bad. Clemson, not ranked and not doing well, is one of the best. What if teams rarely kick? You'd have to have some chance of winning to lose some. Bad offenses that don't get in the end zone much (or close enough to kick) and just punt isn't counted against them. It seems like you'd have to look at this with red zone trips, scoring in the red zone, and percentage of scores that are touchdowns. There just seem like there are so many flaws to this, and as much as it would be nice to have more ammo against Fuente, I don't feel like this is it. I want to see average points scored in the red zone, maybe percentage success in scoring TD when getting the ball first down on the 1. Those are some that I figure would be much more damning to the coaches.

Doesn't that just say that Alabama was been great despite bad coaching decisions?

Free Hugh

Did you read all of what I wrote? All the possible things wrong with it in the first place? It's hard to say that Alabama has been great DESPITE bad coaching. Really? I'm sure they've lucked their way to the national championship game almost every year DESPITE bad coaching. I'm more inclined to think the methodology is at least somewhat flawed. Hell, he even admitted that calling timeouts in certain situations were throwing things off, so yes, let's take this information 100% as bad coaching.

You ok?

Free Hugh

You picked one thing out of everything I said, ignored the rest, and still attributed these graphs to bad coaching. I tried explaining. Again. You ignore that and just ask if I'm okay. I'm fine. You are apparently 100% sold that these graphs mean bad coaching and I'm not going to change your mind.

I picked one to create a simple concise reply. I think the logic stems to all of your examples. What major factors affect a games outcome? Player talent (including development), preparation, and in game coaching decisions have to be 3 of the top.

I picked Alabama because they're the easiest. I think that perhaps they have had the talent and preparation to overcome poor in game coaching decision related to kicking. Not luck. And I would extend that out to all the teams in this list.

Clemson no longer has the talent at the most important position on the field so losing several % points of WP due to coaching choices really really hurts. Possibly to the point where they can't overcome it.

I'm not going to waste my time and type out every single team, but just because you think the methodology is flawed (which it might be idk) doesn't mean I can't say it might be accurate and extrapolate it's significance if it is.

Yes it's not all encompassing of bad teams, but it is still measuring lost WP points which is significant. A coach is making a game decision that is lowering the chance of winning. You could do a similar thing for decisions to go for 2 and you would see our actions decreased our WP there. So I don't see your point.

This isn't meant to be a "you win or lose based on this" this is a "your coach isn't maximizing your chance of winning in these specific situations" stat. There are obviously countless other factors still at play.

I asked if you were ok because you seemed disproportionately agitated to my genuine question/feeling.

Free Hugh

I'm still not buying that you can use this to say bad coaching. Is this averaged across games or total number of plays? If a team scores almost exclusively touchdowns but kicks once (not likely, but hypothetically, but possibly likely in some of Alabama's games), does losing a certain amount of lost WP count more because there was no average? Also, if a team is heavily favored in all games, or is predicted to lose by a lot, how much lost WP will you really get by kicking? To me there just seems to be too many potential flaws (without being able to read methodology) to be able to indict coaching. And even to your point of choosing Alabama to illustrate this, given their winning percentage, I find it even harder to believe this demonstrates poor coaching (I'd look to FSU and how much talent they draw and their performance over the past few years as a demonstration where you can't mask bad coaching with talent). My sarcasm was due to you picking one team that I find it very hard to justify a diagnosis of bad coaching, and ignoring all the possible ways these graphs could be flawed. I think there are plenty of other metrics that much more clearly demonstrate bad coaching, but one where calling timeouts on 4th down somehow changed winning percentage lost is not one I'm willing to put much stock in.

I would lodge an educated guess that the two FG decisions in the A&M game dramatically increased, or accounted for a huge chunk of their WP loss on this graph. On the Cover 3 pod I think Bud mentioned that one of those decisions to kick a FG was the single most costly decision in terms win probability decrease of the weekend.

Drowning in minutia here!

Even when you get skunked; fishing never lets you down. 🎣

Don't worry, I am sure Corny and Fuente's prevent offense will cost us many more wins over the next 4 years till the extension runs out...

I can imagine no more rewarding a career. And any man who may be asked in this century what he did to make his life worthwhile, I think can respond with a good deal of pride and satisfaction:
“I served in the United States Navy"

Scared Money makes money.

I've looked at this graph a half dozen times- and I still have no idea what it measures, whether it is based on subjective or objective inputs, or whether it has any significant correlation to actual success or failure.

So am I just stupid and completely missing it or is this simply just "Fu-man bad, I like it"?

My interpretation:

It's based off something like Win Probability charts Linked Here

Let's say we are at X% Win Probability. But then we decide to kick it in a situation where analytics says "Go", the Win Probability changes from X% to X-7%.

It's based off the likelihood of a team winning based on that game situation, and how a specific decision or outcome changes the likelihood of winning. In this case, Fuente's conservatism on 4th down by kicking/punting costs the team points based on what would likely happen had they not kicked.

The problem with using that is it assumes, say, every 4&1 from the 22 is the same - and they simply aren't. There are countless variables and counter variables.

This seems to me like an utterly useless statistic for real football. But good for fans looking for something to support their own confirmation bias (depending upon their team's current state of affairs).

Sure, but if you have countless football games to use as a guide, and teams that do one thing tend to win while teams that do another tend to lose, that has some level of relevance.

On one hand, you have hundreds of thousands of data points encompassing all manner of situations. On the other hand, you have the gut instinct of someone who thought Malik Willis wasn't good enough to be QB but Ryan Willis was.

Not really.

Sometimes picking the points that "prove" your point don't prove your point.

Sometimes people write things that seem profound but are completely devoid of thought.

Expected Points Added/Win Probability Added doesn't "pick" points at all. It takes a data set of things that have already happened and uses that to predict future likelihood of success. Based on historical data it's determined that Fuente has left points on the field this year and harmed VT's chance of winning due to a lack of aggression. There are certainly limits to the value of such metrics. This does not make them meaningless. Maybe it won't hurt if I drop an anvil on my foot, but historically it hurts a lot when you drop an anvil on your foot, so why is my foot somehow different?

It actually doesn't predict anything.

Nobody here can even explain what it is.

Dropping an anvil on your foot isn't at all like trying to derive a predictive model from a bunch of historical data points.

"It's based off the likelihood of a team winning based on that game situation, and how a specific decision or outcome changes the likelihood of winning. "- Me, explaining what it is

If you don't like my explanation, there's lots of other explanations. There's also a thing called "Google" where you can find explainers from the people who pioneered the idea. Here's one for you.

https://web.archive.org/web/20141215025201/http://www.advancedfootballan...

There's an idea, and there's the implementation of an idea. Those are two different things.

Also, there's facts, and there's the daily cherry-picking through the facts to "prove" what one has already decided.

I don't have to tell you which I believe this is.

There's a lot of valid criticism of Fuente. And there is some that's pretty marginal. I haven't seen anything here to make me think this is towards the valid end of the spectrum.

If Win Probability were comparing VT's 4th-and-short decisions to intentionally selected teams that all happened to be good, It would be cherry-picking. Instead it's comparing VT's 4th-and-short decisions to, well, the entire data set of college football over the years.

You are, of course, free to find metrics that say Fuente's doing a good job and adding to VT's win expectancy through his coaching decisions. Hopefully it doesn't involve calling timeout before the other team attempts a 55 yard FG with 8 seconds left. (See, that's cherry-picking.)

Instead it's comparing VT's 4th-and-short decisions to, well, the entire data set of college football over the years.

Before you get all insulting, perhaps you should try and understand the numbers you're dying on the hill over.

The graph says plainly that these numbers are for the first six games of 2021.

Also, there's this, from the posting itself:

This is awkward, my apologies to Virginia Tech, there was a bug that was counting timeouts on 4th downs as lost WP. Poor work by me there. Virginia Tech was most affected because they took a timeout immediately before they made their worst 4th down decision of the season

I'm not pretending to be a statistician, but like I said earlier, there are plenty of valid criticisms of Fuente that don't venture this far from defensible statistical reasoning.

I think that it's fair to bring this posting up for discussion, as it's rare to be that far outside the bell curve, and it's been presented publicly. I just don't know how valid a data point it really is.

Can you explain why we're the outlier and not somewhere within the bell curve?

Free Hugh

It's an Obscure Statistical breakdown nobody has ever brought up before on this site with the handy acronym "WPPGLBKIGS".

And that's precisely where you find outliers for teams with a 3-2 record.

I don't know about you, but phonetically, "Whupgulbakigs" really rolls off the tongue

21st century QBs Undefeated vs UVA:
MV7, MV5, LT3, Braxton Burmeister, Ryan Willis, Josh Jackson, Jerod Evans, Michael Brewer, Tyrod Taylor, Sean Glennon, and Grant Noel. That's right, UVA. You couldn't beat Grant Noel.

sounds German

đŸ”„â›Č

Rolls off the tongue like Dutch.

This joke gets my Begbie seal of approval

Fu's decisions in this area are clouded by the fact his kicker's accurate range is around 35 yards. That fact skews your decision making vs. if you had a justin tucker for example.

I agree but also think that's a bit of a stretch considering he had just made a 50+ yarder before half. I know he missed against wvu from in tight, but you gotta trust your guys right?

Oh wait, not if you're CJF

VB born, class of '14

See the comment directly below yours for some context.

The sad part is he is less accurate within 35 yards and more accurate over 35.

Can someone explain to me what this is actually measuring?
Is it saying the % of games lost because we kicked in goal to go situations?

VT 2016
Go Hokies

See my post a few above yours. At the very least, I'm not the only stupid one.

I'm curious to see if anyone can actually explain what it measures and whether it's built on objective or subjective measures.

Book vs. gut, and how it effects your win probability.

We go for it and get the 4th and 1, we still have a greater than 50% chance to win the game. But we kick instead, and gave the ball back to ND, and our chances of winning fell to nearly 0%.

So we gave up almost 60% chance of winning the game. That's what the bar is telling you.

The graphs tell you how this has occured over 6 weeks of play. BC plays it by the book. They don't go for it on 4th down a lot. But when they kick instead of going for it, it does not effect whether they win or not.

Illinois and UVA are 2 and 3 respectively. That means they've given up on games by kicking instead of going for it on 4th down. They are giving away win probability.

We have lost 2x. But in both games, we had a greater than 60% of winning the game. Instead of going for it late in games, we have kicked, taking away our probability of winning. Remember, we were going to lose to WVU if Waller doesn't make a play b/c we kicked the ball back to WVU (or there was a fumble, don't remember which happened first).

Against UNC, we kicked the ball back, before our defense made the game winning interception. We were giving UNC a chance, and hoping the defense could stop them. This lowered our win probability.

TLDR: the graph is telling you how much we're willing to give up on a game, and take a lottery chance at 1. getting the ball back (thank you Waller), or 2. the opponent not scoring (fuck you ND)

TKPhi Damn Proud
BSME 2009

I could be wrong but I'm pretty sure this is talking about kicking a FG inside the 10 yard line. The title is not really clear. If it was about punting the ball back to the other team it would say punting and not kicking.

We were not inside the opp 10 yard line when we punted the ball back to ND Saturday for the game winning field goal. That data metric is included in this chart.

TKPhi Damn Proud
BSME 2009

How do you know anything about punting is in here? Was the methodology for this posted somewhere? Because, as far as I can tell, this ONLY measures kicking (a specific term that is not punting) and not even kicking in all situations, only in situations where a team is goal to go.

Its literally in the guys profile. See my post below.

(add if applicable) /s

I'm just going to put this here cause there are a lot of people saying they don't understand and multiple threads about what is or isn't included and 0 people that actually took the time to look at the guys profile and find the source of the data.....

https://kazink.shinyapps.io/cfb_fourth_down/

(add if applicable) /s

yes "go situations" was confusing for me as i processed it as "goal to go situations".

"Why gobble gobble chumps asks such good questions, I will never know." - TheFifthFuller

Yeah he could update the headers its a 4th down go/no go situation and statistically what you should do vs what was done and how that decision hurts your likely hood of winning.

VTs Punt against ND was the single worst decision made this season in P5 in terms of this metric.

(add if applicable) /s

I think you're right about the "go situation" - see my post below about changing the inputs. The page you get to aren't the results used in the graph, it's the results of the current inputs in the model, which you can change to any game situation.

Okay, did you open the inputs and change them, or did you just look at what showed as the output? Because what showed in the output is based on the settings currently in there, which are 4th and 5 from the opponent 35 yard line. If you click on the plus sign by inputs, you can change all the settings. "Going for it" as shown in the output table is not the same thing as being in a "go situation" (goal to go). If you change the inputs (like I did) to being less than 10 yards, it puts "NA" on the punt line, which looks a lot like punting is NOT included in the metric shown in the original post.

Edit: the model you linked to is showing the correct result for the current inputs. It's saying that if you're 4th and 5 on the opponent 35 in the 4th quarter with 3:30 on the clock and you're tied, the correct thing to do is punt given the odds of success of going for it or kicking a field goal (both low in that situation, so you should punt). Change the inputs to different scenarios and it will give you different results.

That's not where this information is derived from, that's a model situation calculator you can mess around with.

Check the inputs data in the About section or look at VT in 4th Down Plays - its a single page application and the model calculator is the landing page, I can't link to those sections.

Edit: the model you linked to is showing the correct result for the current inputs. It's saying that if you're 4th and 5 on the opponent 35 in the 4th quarter with 3:30 on the clock and you're tied, the correct thing to do is punt given the odds of success of going for it or kicking a field goal (both low in that situation, so you should punt). Change the inputs to different scenarios and it will give you different results.

I have 0 idea what you're referring to here. Every 4th down play VT has had again P5 oppenents is already broken down you don't need to use that calculator.

(add if applicable) /s

What shows on the 4th Down Plays are actual game results. It shows one team that had 9 yards to convert a 1st down with 64 yards to go to goal. It says the ideal choice is a punt and the actual choice made was to punt. That tab is NOT the result of an analysis on winning percentages lost by kicking because it shows plays where the ideal situation was to go for it and teams actually went for it. If that tab was showing winning percentage lost in go situations, there wouldn't be any plays showing where teams went for it.

Edit: My point being there's a model in the link and a lot of data showing on 4th down plays. The 4th down plays show results for going for it, so are not strictly what was used in calculating the results behind the graph above because those explicitly say loss for kicking. You argued that punting is included, and just because punting is shown in the link, doesn't mean it was included in the calculations to get to the results above, because we know for sure going for it isn't included in the results above (by definition not a kick), yet is also included in the results in the link.

The chart clearly says "WP lost by kicking" at the top of the chart.

TKPhi Damn Proud
BSME 2009

I know - Heuplek said it included punting, and I'm trying to say punting is not factored into "kicking" and even though it shows punting in the link provided, it also shows going for it, which definitely isn't a kick, therefore just because punting results are shown on the link doesn't mean it's included as a "kick." My understanding is that the guy who calculated this only used results where a team attempted a field goal (strict definition of kicking in football) and then only in situations where it was goal to go and corresponding loss in WP.

From the original tweet:

Looking at WP lost per game on 4th down decisions for P5 schools this season according to
@aisports_4th

The entire statistic is based on 4th down there's nothing about goal to go anywhere. We can literally set the filters and see our scenarios based on his data for "GO Situations" (from the chart) which is an included filter on the data.

You're starting to place doubt in my head but I don't see why he would just randomly go off the wall when he has all of this on his website...regardless of who's right the language in the chart is awful the only way I get to his number in the graphic is only punts on statistical 'Go Situations'

4th down FGs only:

4th Down FGs and Punt
And to really throw a wrench in everything the only way to get to the -6.8% number in the graphic is to actually remove FGs all together and average the games out.

(add if applicable) /s

I meant to write back yesterday, but had actual work I needed to get done (it's a sad day when my paycheck gets in the way of me trying to TKP harder). Anyway - I'm confused too. I tried filtering that almost every possible way, and the only way I came close was to limit it to 4th quarter, plays that should be "go", but instead punt or kick a field goal. I got -7.1. But considering the title says "per game" that doesn't seem right. I just couldn't get it to add up. I also noticed in the list of opponents or "all opponents" it didn't list Richmond or Middle Tennessee, so not sure if those were excluded from the analysis, or we just can't recreate it because they were using a larger data set than what they're showing. Either way, or any way, I'm still not convinced you can definitely read this stat as bad coaching. I still think we have bad coaches (pretty much exclusively on the O side of things (not that they're all bad, just the bad ones are there)), so not going to use this as a data point for badness of coaching. And absolutely agree the language is bad - and for me to feel this is valid, we should easily be able to recreate his findings).

I'm glad you didn't because I spent like 3 hours trying to get all of the numbers to match the graphics using his raw data set and couldn't get it. Using the same method on multiple teams I could match the graphic about 60% of the time then suddenly a team would be ridiculously far off.

I'm now embarrassed that I called out everyone for not looking at the background data, then looking at the background data and getting even more confused, so apologies there. I'm gonna have to tweet at this guy if he posts this again this week because it makes 0 sense how he's deriving these numbers.

(add if applicable) /s

You're good dude. Honest mistake. I agree - the data he has should back up the graphs. Not sure anybody else was interested in our conversation, but I'm a number cruncher, so I'm glad you challenged so we could try to validate his results. I was dubious of the validity at first not because of the numbers he would have used, but more because of the other related issues I've brought up (like a team that was crushing their opponent and kicked would likely lose less WP than a team that had a close game and was narrowly favored to win and kicked in a "go" situation). I mean if I was bored, I could probably come up with all sorts of stats that sound good, but don't tell you anything really useful.

Just in case anybody is forgetting, getting inside the 10 and having a "go situation" does not automatically mean 10 yards or less - if you have a penalty of the 5, 10, or 15 yard variety, and/or sacks, you can be looking at a lot more than 10 yards. I distinctly remember our defense getting some sacks on opponents and them losing major yardage - I think I remember between penalties and sacks backing up at least on opponent to the 30. If that happened, say penalty, sack, and an incomplete pass and a team is on the 20 or 30 on 4th down and they're still in a goal to go situation, then they kick a field goal, does it count lost winning percentage on those 4 downs, or just on the actual play where they kicked rather than went for it? So is it bad coaching if the defense is able to sack your QB a couple times and forces an incomplete pass and you decide to kick? You can argue that your team getting penalties and putting you far enough back that the team has to go 20 yards to get to the goal but can't do it so you kick a field goal is bad coaching, but probably more from a teach them not to commit penalties perspective and not on the decision to kick.

That, plus the things I brought up in comments above (such as whether a team scores touchdowns on 95% of go situations, kicks in those situations where the winning percentage was already incredibly high or low, your offense being so bad a team rarely gets in goal-to-go, or so good they score more often than not from further away and rarely even downed inside the 10, etc., etc., etc.) make this feel even more like a metric some statistics nerd came up with then applied a graph to in order to make a splash, but not very helpful. Hell, considering Tech is the worst team, maybe he's a LOLUVA grad who was able to create some metric just for the point of us being last (hence the LOL VA Tech line in the beginning).

Also, let me point out again that calling timeouts on 4th down somehow messed with the numbers - which would cause me to question the validity of other factors that were built into this POS. As I said above, there's lots of stuff that makes our coaches look bad, I just think this isn't one of them.

The entire raw data set is available. My guess is timeouts aren't factored but simply rolled into the data set from his extract and weren't ignored which would obviously cause errors.

(add if applicable) /s

Given that UVa is SECOND to last, doesn't seem like it was created by a UVa nerd, at least not a smart one.

Edit: didn't mean to be insulting.

Was just reacting on the UVa nerd part.

The rest of the post I agreed with.

Doesn't sound insulting to me. It's a good point. I should have looked more closely at that. It could be that certain UVA grads don't care about being bad though as long as we're worse, especially if we are THE worst. Not sure why the LOL VA Tech in the original post - that's just shitty.

Edit: from his Twitter- he's a Ute, so just LOL'ing at us for nothing other than being the worst team in his made up statistic. Also not sure why someone would downvote you on this - was it you that had the serial downvoter? Leg to you for keeping me straight and to counter the hater lol.

I said it in another thread - "Good teams find ways to win, bad teams find ways to lose" - this is statistical evidence of that adage.

Twitter me