Charger Coordinator/Staffing News and Discussion

[Critty]

Really like that they want to use analytics as part of things.

At the same time a players desire to be great cannot be capture with analytics.

And with analytics, how far do they take it.
Is it game plan down and distance analytics type of things.

Are they trusting someone else's grade like PFF.

And if they do much of their own in house, how far do they take the analytics. Do they track each and every play by a player and put together a chart based on some sort of criteria.

For instance, how does analytics track a TE who not only execute his block but put the linebacker on skates and drove him 10 yards downfield.

If they have stats of block executed by TE.
How would they do it.

TE on 230-250lb Linebackers likely to be draft high, and how many yard driven and was there a pancake.

TE on 230-250lb linebackers likely to be drafted low or udfa, and how many yards driven and was there a pancake

TE on 230-250lb linebackers who will not get drafted and unlikely to make a roster, how many yards driven and was there a pancake

And do they just keep going TE vs SS vs FS vs Edge with what criteria.

And the TE on pass route vs what criteria. Etc etc.

Also add in size and speed and do they also account for awareness and great hands. Doesn't fumble. Doesnt make penalty.

And after all that start to rank the TEs for the 2021 draft.

Or do they just go this TE didn't do well in many of these things above, but he is clearly a weapon and a 1st.rd prospect.

The only place to find and build that info is film study and very detailed notes of each and every play.

[sonorajim]

Analytics is a tool.
Staley using it along with his own observations, instincts and experience along with that of his staff, where appropriate may be an art form. He's reputed to be a genius.
I hope he doesn't simply regard analytics as The Book with no personal input required in it's application.

[Boltdiehard]

Analytics is a tool.
Staley using it along with his own observations, instincts and experience along with that of his staff, where appropriate may be an art form. He's reputed to be a genius.
I hope he doesn't simply regard analytics as The Book with no personal input required in it's application.

Agreed and it also helps to have the right people and the Tampa Bay Rays are a good example.

[Boltjolt]

Agreed and it also helps to have the right people and the Tampa Bay Rays are a good example.

Assume you meant the TB Bucs.

Just want to point out that it was Brady who wanted Gronk and AB. Arians didn't want to sign AB till Brady insisted.

[Boltdiehard]

Assume you meant the TB Bucs.

Just want to point out that it was Brady who wanted Gronk and AB. Arians didn't want to sign AB till Brady insisted.

No I meant the Rays as far as analytics go. Different sport I know but they rely heavily on analytics and it works for them and quite well. Buncha whizkidz in that front office.

[Xenos]

Really like that they want to use analytics as part of things.

At the same time a players desire to be great cannot be capture with analytics.

And with analytics, how far do they take it.
Is it game plan down and distance analytics type of things.

Are they trusting someone else's grade like PFF.

And if they do much of their own in house, how far do they take the analytics. Do they track each and every play by a player and put together a chart based on some sort of criteria.

For instance, how does analytics track a TE who not only execute his block but put the linebacker on skates and drove him 10 yards downfield.

If they have stats of block executed by TE.
How would they do it.

TE on 230-250lb Linebackers likely to be draft high, and how many yard driven and was there a pancake.

TE on 230-250lb linebackers likely to be drafted low or udfa, and how many yards driven and was there a pancake

TE on 230-250lb linebackers who will not get drafted and unlikely to make a roster, how many yards driven and was there a pancake

And do they just keep going TE vs SS vs FS vs Edge with what criteria.

And the TE on pass route vs what criteria. Etc etc.

Also add in size and speed and do they also account for awareness and great hands. Doesn't fumble. Doesnt make penalty.

And after all that start to rank the TEs for the 2021 draft.

Or do they just go this TE didn't do well in many of these things above, but he is clearly a weapon and a 1st.rd prospect.

The only place to find and build that info is film study and very detailed notes of each and every play.

Currently I’m thinking that Staley would be using analytics more on the coaching side. But it would nice to start using it more towards scouting and the draft. Maybe not this year but probably starting next IMO.

In terms of how it could be implemented this year, I was thinking of what Reich and the Colts did, and also Harbaugh and the Ravens. Let me know if you can read the articles below:

https://theathletic.com/549166/2018/09/26/how-analytics-are-changing-the-colts-one-decision-at-a-time/

https://theathletic.com/1338133/2019/10/31/evolution-of-aggression-how-colts-coach-frank-reich-became-one-of-the-games-best-play-callers/

https://theathletic.com/1396091/2019/11/22/analytical-edge-how-john-harbaugh-and-ravens-have-gained-an-advantage-with-fourth-down-aggressiveness/

https://theathletic.com/2321294/2021/01/14/colts-coach-frank-reichs-use-of-analytics-is-here-to-stay/

[SBbound]

. Let me know if you can read the articles below:

I cannot. Probably time for a subscription.

[Xenos]

I cannot. Probably time for a subscription.

When I have time, I can post them one by one. But if you are getting a subscription, make sure to look for the discounts. I’ve been using the 50% ones that I occasionally see.

[Bolt4Knob]

I cannot. Probably time for a subscription.

go to Twitter and ask Popper if they are running any specials - seriously
even copy Jourdan Rodrique

[SBbound]

go to Twitter and ask Popper if they are running any specials - seriously
even copy Jourdan Rodrique

Requested, thanks.

[FoutsFan]

Analytics is a tool.
Staley using it along with his own observations, instincts and experience along with that of his staff, where appropriate may be an art form. He's reputed to be a genius.
I hope he doesn't simply regard analytics as The Book with no personal input required in it's application.

It took Dave Roberts of the Dodgers several years of being stabbed in the back by his analytic team to figure that out. His Excel spreadsheet would manage the Dodgers and Roberts would just be the mouthpiece for the spreadsheet. Once Roberts figured that out last year and he used the spreadsheet, not the other way around he started to be a good manager.

Analytics are great but you cannot let it replace human understanding and intuition.

[Xenos]

sonorajim and FoutsFan this article provides more of what I'm thinking about with regards to analytics in coaching. Analytics cannot determine the right play to use (at least not yet). It's best usage so far is determining binary choices ie. going for it on 4th down versus punting, or going for a TD versus a field goal. The infamous Frank Reich's decisions in the recent Colts vs Bills playoff game is the example used.

https://theathletic.com/2321294/2021/01/14/colts-coach-frank-reichs-use-of-analytics-is-here-to-stay/

As Frank Frigo watched the late-first half sequence of events in the Colts-Bills AFC wild-card game on Saturday, he made a prescient comment to a colleague as the final seconds ticked away.

Frigo, co-founder of sports data firm EdjSports, knew what was coming.

“I made the comment to one of our analysts, ‘What do you think they’re going to talk about more during the halftime show? The fact that (the Colts) failed during the fourth and 4 or that there was a crucial, almost unnecessary offsides penalty?’ ” Frigo recalled this week.

Sure enough, the CBS studio crew immediately turned its focus to the debate surrounding Colts coach Frank Reich’s unsuccessful fourth down from the Bills’ 4-yard line. They did so while never mentioning a subsequent play that, in Frigo’s data-based opinion, had a much greater influence on the Colts’ loss than their failure to score a few minutes earlier.

Defensive end Kemoko Turay’s inexplicable offsides penalty on a fourth down with 37 seconds remaining, when the Bills were actively trying to entice the Colts to jump to get a cheap penalty, set up Buffalo’s go-ahead touchdown three plays later. That translated to a four-point swing considering the Bills were content to call a timeout and attempt a 43-yard field goal before the penalty.

“This is a perfect example of how we have selective memories,” Frigo said. “No one talked at the half about that penalty and about how, the moment it occurred, it changed the game state. It went from a fourth and 3 to a first and 10, and that boosted the Bills’ win percentage by 7.5 percent. Which is huge.

“So, here we have a decision that we thought was a pick ‘em decision that Reich made. We think there was about a half-percent of win probability in the balance. They fail and everybody remembers that but nobody talks about the offsides penalty that cost them 7.5 percent.”

Frigo is correct. Then again, this is the path that Reich has chosen. He has bought into analytics more than most NFL coaches, which can have very positive results over the course of a season if you believe the data. But if you adhere to the numbers even when they suggest actions that seem unconventional, you’d best be prepared to be second-guessed.

Reich drew significant criticism for his decisions in the final minutes of the half, a crucial juncture in the Colts’ eventual 27-24 loss.

There are two separate issues at play: There is decision-making, like Reich’s election to attempt the fourth down. Then there is play calling, which is completely separate (i.e. the perimeter run Reich called on third down that resulted in a 3-yard loss).

Reich took shots for his calculations in both areas.

Frigo, a former backgammon world champion who spent two decades working in commodities markets, is no expert on play calling. But he certainly has thoughts about decisions like fourth downs and onside kicks. The way he saw it, Reich was not even a little bit reckless by opting to forgo the field goal and attempt the fourth down with his team up, 10-7.

EdjSports (pronounced edge) has a sophisticated model that has been developed over the years. And when simulating the Colts’ fourth-down scenario, the model suggested attempting the fourth down improved the Colts’ “game-winning chance” by 0.5 percent. As Frigo said when explaining his reaction to Turay’s blunder, that mistake negatively impacted the Colts’ game-winning chance by 7.5 percent. And that’s to say nothing of the resulting touchdown.

Different models can produce different results because their methodologies can vary. The Athletic’s Ben Baldwin, who writes about football analytics, has his own model. That model favored kicking a field goal on fourth down by 2 percent – still a slim margin. Additionally, Baldwin’s model put the Colts’ game-winning chance at 60 percent had they scored a touchdown and 51 percent with a made field goal.

The Colts use several data services and those companies’ models all slightly favored a fourth-down attempt. Each of those models also put the Colts’ win probability north of 70 percent by scoring a potential touchdown in that situation.

Reich will have to live with his play selection during the long offseason. The third-down toss to Jonathan Taylor sounded good in theory, but it is hard, in retrospect, to argue against running the ball up the middle behind dominant linemen like Quenton Nelson and Ryan Kelly.

“We hadn’t tossed it all year in that scenario,” Reich said. “It breaks a tendency, but the easier call would have been run it up the middle behind our great offensive line. But I didn’t make that call. I look back on that now because the other one didn’t work, yeah, I wish I would have done it.”

Play-calling decisions are much more complex because they must consider variables like tendencies, as Reich laid out, or elements like your team’s strengths and those of the opposing offense or defense.

But when it comes to in-game decisions where there are limited options – kick a field goal or attempt a fourth down, timeout usage, etc. – data can simplify the choice. And Reich is committed on this front. Frigo has met with Reich about data use and even communicated with Colts general manager Chris Ballard after Saturday’s game, so he is quite familiar with the team’s approach. Colts analytics staffers, George Li and John Park, regularly communicate with Reich during games to assess upcoming decisions. Frigo sees substantive differences between Reich and other coaches that set him apart in this area.

“What I liked about him from Day 1 is that he takes a very unapologetic approach to it,” Frigo said. “He gets the idea that you make good decisions and sometimes they work and sometimes they don’t. One thing that always makes me cringe is when you see a coach make a really solid, analytically-based decision, but if it doesn’t work, they go back and get all ultra-conservative the next time. Why does it matter? Sometimes it works, sometimes it doesn’t. But you have to do the right thing. In the long haul, you’re going to benefit from it.

“But Reich is very much like, ‘I’m following the math and over the long haul we’re going to do well because of it.’ He has been really impressive.”

Reich was ranked EdjSports’ top decision-maker in the 2020 season. That means he topped its “critical call index,” which tracks things like consequential fourth-down decisions (it weeds out fourth downs in the final minutes of a game when a team is trailing).

It’s difficult to assess how much this approach impacts a team’s results. So many other variables factor into wins and losses, like talent. Making sense of it all is hard. But there is some anecdotal evidence that shows Reich’s use of data is having positive results. The Colts, for example, have improved their fourth-down conversion rates every season under Reich while also attempting more fourth downs each season.

In 2018, the Colts converted 46.7 percent of their 15 fourth downs. In 2019, the number jumped to 58.3 percent on 24 attempts. And in 2020 the Colts were successful on 61.5 percent of their 26 attempts.

Could that be an indicator of Reich and his staff growing more comfortable in their use of data? Perhaps. If so, that’s a very good thing; fourth-down conversions have a significant impact on game-winning chances in predictive models because teams retain possession which provides another scoring opportunity.

That’s why Frigo and most analysts excoriated Titans coach Mike Vrabel for punting from the Ravens’ 40-yard line while trailing early in the fourth quarter of Sunday’s AFC wild-card game (EdjSports estimated the decision cost Tennessee about 14 percent in game-winning chance). That’s certainly an extreme example, but Frigo says he often hears from coaches who argue against attempting borderline fourth downs when, say, their offense might be struggling or their defense is having a particularly strong performance.

“What they’re really saying is, ‘We were trailing and I wouldn’t go for it, so what we did was punt the ball away, trusted our defense that they were going to get a stop, so I could then give the ball back to that same offense that I had no confidence in,’ ” Frigo said. “Believe me. This happens all the time.”

Even if you can appreciate what Frigo espouses, you might still read this story with some trepidation. Trusting data often requires us to put aside some long-held beliefs about football that are hard to ignore.

That can even be difficult for NFL coaches.

“There’s still a long way to go,” Frigo said. “Analytics suggests some very counterintuitive ideas. The decisions that have huge regret when they fail are still eluding most NFL coaches. It’s most often very, very correct to go for it in your own territory on fourth and 1. It looks disastrous when you fail. You immediately hand your opponent possession in the red zone.

“But the thing the model is picking up on is that you’re a favorite to convert those. When you do, it’s a huge benefit. And when you fail, you are going to probably be punting from a disadvantageous position anyway. It’s amazing how many times we see teams punt on fourth and short backed up in their own territory and then, like, two plays later, the opponent is already on their 35-yard line. And now they’ve also eaten a bunch of clock.”

It’s a lot to digest. Much of this is a bridge too far even for Reich. But there’s some compelling data about the impact of ignoring the numbers. Frigo has seen coaches yield as much as 1.5 games in cumulative game-winning chance during an entire season of questionable decisions.

On the other end of the spectrum, how can we determine that the teams and coaches committed to analytics are benefiting?

“I can tell you that three of the highest-rated analytics teams are sitting in the playoffs right now,” Frigo said. “Watch the way those teams behave.”

He points specifically to the Packers’ Matt LaFleur, the Ravens’ John Harbaugh and the Chiefs’ Andy Reid as three prime examples.

“There’s definitely the haves and the have-nots,” Frigo said, “and you’re starting to see the cream rise to the top.”

Will Reich rise along with them? If he’s right about his embrace of data, he just might. But we can be sure of this much: Reich’s reliance on analytics isn’t going away anytime soon.