Browns hire Jonah Hill....err Paul DePodesta as CSO
-
What? I really don't understand what you're saying. Look, I don't look at the models these guys do. but for example I'm quite certain if the game is tied 0-0 in the first quarter and your team is facing a 4th and 20 from your own 5, the model would spit out a 99.9999% opinion to punt the ball. That would qualify in stats as "common sense"
I work with stats every day albeit in a way, way different field. I am biased which I admit, but I see where stats can heavily inform opinions in a positive manner all the time. It's crazy and doesn't make sense even to me sometimes, but we admit students that on paper shouldn't be in the program but excel. Who knows what Haslam is really thinking hiring these guys but you should resign yourself to the idea that stats and analysis will play heavily into decisions. Of course stats can be wrong, but the Browns certainly haven't gotten much right up until today right? Of course this stuff can't be the only way to make decisions, but it can really inform decisions. It'll be very, very interesting to see how it plays out because if the Brownies finally get a break it could be the start of a new wave of thinking in the game.
I agree with you in that common sense would say to punt, but that's not really the way it's figured in this point of view. It's more of an expected points situation. And that's where things get wacky, but yet also sort of... make sense?
In the most basic of rationality, the further in your own territory you are, the higher chance the other team has to score off of your punt. For instance, punting from your own 5 would give the opposing team better field position than, say, punting from your own 35. Better field position then leads to a higher expected point value for the other team.
Add in the fact that punting yields an expected return of 0 points for you, where going for it yields an expected return of > 0 points, and you can see why the case can be made to never punt...and why it would actually make less sense to punt the further in your territory you get backed up.
This is just one situation of how analytics can be completely skewed.
Of course, common sense would dictate that you punt in that situation. But looking at the expected point values, it would actually make more sense to go for it. That's because most of these models (at least the ones that I've read so far) don't account for turnovers as negative points.
I agree with you in that common sense would say to punt, but that's not really the way it's figured in this point of view. It's more of an expected points situation. And that's where things get wacky, but yet also sort of... make sense?
In the most basic of rationality, the further in your own territory you are, the higher chance the other team has to score off of your punt. For instance, punting from your own 5 would give the opposing team better field position than, say, punting from your own 35. Better field position then leads to a higher expected point value for the other team.
Add in the fact that punting yields an expected return of 0 points for you, where going for it yields an expected return of > 0 points, and you can see why the case can be made to never punt...and why it would actually make less sense to punt the further in your territory you get backed up.
This is just one situation of how analytics can be completely skewed.
Of course, common sense would dictate that you punt in that situation. But looking at the expected point values, it would actually make more sense to go for it. That's because most of these models (at least the ones that I've read so far) don't account for turnovers as negative points.
Isnt the analytics going to be used in player acquisition. The coach will coach the players and make game decisions. How analytics will translate into player acquisition I have no idea, I would have to see what metrics they would be using to help them make decisions. Baseball is a stat sport even before sabre metrics came out so it makes a lot of since there. I think its much more difficult for football.
-
-
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
-
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked for pretty well for George Costanza.
We should have kept Farmer and Pettine on board - especially since we're still paying them.
Just let them finish their big board and work away from that.
-
Of course, common sense would dictate that you punt in that situation. But looking at the expected point values, it would actually make more sense to go for it. That's because most of these models (at least the ones that I've read so far) don't account for turnovers as negative points.
Interesting, that would be really odd to me that turnovers wouldn't factor in a negative expected points value. I really don't know anything about the models they use though-I get that at a base level but it seems the model could account for that.
Interesting, that would be really odd to me that turnovers wouldn't factor in a negative expected points value. I really don't know anything about the models they use though-I get that at a base level but it seems the model could account for that.
In the couple studies I've found thus far, the running theme is either "you score or you don't score", where each individual possession is looked at in a bubble, rather than in a linear model.
In that line of thought, turning the ball over is, in effect, the same as punting...in that you didn't score. You're just giving the opposition the ball in a much better position for them to score. So a 3-yard run on 4th and 10 is no different than a 3 yard punt, basically.
It's wacky. Like I said, it kind of makes sense...but it's also just wrong because football is a causal sport. Every single thing that happens in a football game until the first score is based off of what happens on the opening kick. Then everything until the next score is based off of the next kick. It's not like baseball, where you get a completely clean slate the next inning (if nobody scores).
I'm awesome, and I fuck dolphins. Ghoolie 12/2/2012: I have absolutely no skills, training or pride -
Banner did give a good example of analytics potential conflict with traditional coaching philosophy: passing early vs. establishing the run.
Analytics say the team leading at the half has the greater chance to win the game and passing leads to more scoring therefore you pass early.
Traditionally HCs, and Defensive minded HCs in particular, would resist passing in favor of ball control early setting up the pass game late.
-
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
Which is exactly why I said fuck the QB chase and draft Bosa. Every year the Browns chase QB's with exactly the same results. And right on this board I'm watching Browns fall in love with this or that QB. Last week it was Lynch (until he got his clock cleaned). This week it's Goff. Fuck that shit. Pick the best fucking player available, not the fucking position. It's never worked here. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
-
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
Which is exactly why I said fuck the QB chase and draft Bosa. Every year the Browns chase QB's with exactly the same results. And right on this board I'm watching Browns fall in love with this or that QB. Last week it was Lynch (until he got his clock cleaned). This week it's Goff. Fuck that shit. Pick the best fucking player available, not the fucking position. It's never worked here. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
Which is exactly why I said fuck the QB chase and draft Bosa. Every year the Browns chase QB's with exactly the same results. And right on this board I'm watching Browns fall in love with this or that QB. Last week it was Lynch (until he got his clock cleaned). This week it's Goff. Fuck that shit. Pick the best fucking player available, not the fucking position. It's never worked here. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
We have drafted the consensus #1 QB exactly once in the history of the new Browns. Maybe once in the entire history of the Browns?
I don't disagree with you, but this situation isn't necessarily the same.
Goff + whoever or Bosa + Wentz are the only two options I'd like to see.
I'm awesome, and I fuck dolphins. Ghoolie 12/2/2012: I have absolutely no skills, training or pride -
Hopefully the analytics will simply look at what the Cleveland Browns would normally do, who they would sign and who they would draft............ and completely do the opposite.
Hey, it worked pretty well for George Costanza.
Which is exactly why I said fuck the QB chase and draft Bosa. Every year the Browns chase QB's with exactly the same results. And right on this board I'm watching Browns fall in love with this or that QB. Last week it was Lynch (until he got his clock cleaned). This week it's Goff. Fuck that shit. Pick the best fucking player available, not the fucking position. It's never worked here. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
Two former Browns QB's if we're being honest.
-
. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
Two former Browns QB's if we're being honest.
Two former Browns QB's if we're being honest.
Playoff appearance has little to do with the QBs on that team
-
-
Wondering how Lou Groza would have fit into this analytics thing..
Our musical works to date http://www.arielsatticmusic.com/ -
-
I love it. To be honest it seems like we are doomed to loose no mater what we do. I would love to see us change the game and get ahead of the curve. I just hope that we can get the football people we need at coach and "GM" to buy into it and actually make it work. Our situation is such a mess that we need a major shake up to turn this thing around. I'm willing to give this a shot.
-
I'm pretty stoked about this, and I think this is a ballsy move by a team that is already under tremendous scrutiny. This is not playing it safe, this is not doing what people expect, and this MIGHT be just what we need. We aren't trying to catch up to other teams, we're trying to forge our own path. I think that's pretty cool.
-
Interesting, that would be really odd to me that turnovers wouldn't factor in a negative expected points value. I really don't know anything about the models they use though-I get that at a base level but it seems the model could account for that.
In the couple studies I've found thus far, the running theme is either "you score or you don't score", where each individual possession is looked at in a bubble, rather than in a linear model.
In that line of thought, turning the ball over is, in effect, the same as punting...in that you didn't score. You're just giving the opposition the ball in a much better position for them to score. So a 3-yard run on 4th and 10 is no different than a 3 yard punt, basically.
It's wacky. Like I said, it kind of makes sense...but it's also just wrong because football is a causal sport. Every single thing that happens in a football game until the first score is based off of what happens on the opening kick. Then everything until the next score is based off of the next kick. It's not like baseball, where you get a completely clean slate the next inning (if nobody scores).
In the couple studies I've found thus far, the running theme is either "you score or you don't score", where each individual possession is looked at in a bubble, rather than in a linear model.
In that line of thought, turning the ball over is, in effect, the same as punting...in that you didn't score. You're just giving the opposition the ball in a much better position for them to score. So a 3-yard run on 4th and 10 is no different than a 3 yard punt, basically.
It's wacky. Like I said, it kind of makes sense...but it's also just wrong because football is a causal sport. Every single thing that happens in a football game until the first score is based off of what happens on the opening kick. Then everything until the next score is based off of the next kick. It's not like baseball, where you get a completely clean slate the next inning (if nobody scores).
Yeah, I mean that all makes sense at a base level but I'd be surprised if they couldn't create a model to counteract that for football. Of course, modeling software that does that literally costs $500,000 sometimes, but Haslam can sneeze that much towards the cause.
-
Which is exactly why I said fuck the QB chase and draft Bosa. Every year the Browns chase QB's with exactly the same results. And right on this board I'm watching Browns fall in love with this or that QB. Last week it was Lynch (until he got his clock cleaned). This week it's Goff. Fuck that shit. Pick the best fucking player available, not the fucking position. It's never worked here. Hell we just watched a dominant defensive player pull a former Browns QB right into the playoffs. Ring any bells?
We have drafted the consensus #1 QB exactly once in the history of the new Browns. Maybe once in the entire history of the Browns?
I don't disagree with you, but this situation isn't necessarily the same.
Goff + whoever or Bosa + Wentz are the only two options I'd like to see.
We have drafted the consensus #1 QB exactly once in the history of the new Browns. Maybe once in the entire history of the Browns?
I don't disagree with you, but this situation isn't necessarily the same.
Goff + whoever or Bosa + Wentz are the only two options I'd like to see.
I find myself strangely agreeing. Not so sure on Wentz, but I'm hoping for Bosa and a developmental guy (with McCown being able to start for a year or two) or Goff and then who knows.
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login