I can manipulate statistics to my favor just like the books in business.
Great skill. Math is what you want it to be if you know how to manipulate the numbers (right Enron?)
2 Likes
In their defense
Who really thought anybody would check after the 2nd decimal place to find the fraud ???
1 Like
I see what you did there 
A good accountant can make the same set of numbers say two different things. A great accountant can make the same numbers say whatever you want and have it be true.
Statistics don’t lie, but Statisticians do.
1 Like
64.94563% of all statistics are made up.
1 Like
I think you can simplify this quite a bit.
Tampa was going to try to win in regulation if possible. Hostile environment. On the road. It’s what Campbell did in Dallas. Go for the win on the road.
So let’s assume you score twice, and to make it easy, let’s just say extra points are 100%. And let’s just say 2 point conversions are a 50/50 proposition.
So, if you kick it on the first touchdown, there’s a 50% chance you win if you convert on the two point try, and a 50% chance you lose if you fail to convert. Pretty simple.
Now, if you convert the first, there’s again a 50% chance you win if you convert (because all you need is the touchdown that you have to score anyway, and we’re saying kicks are 100%). But if you don’t convert, you now still have another try. So in that 50% of the time, there’s now half the time where you convert and tie the game and half the time you lose.
So now it’s about 50% win, 25% tie and go to overtime, and 25% lose.
Now you can break it down further into chance you think you win in overtime, etc, but you can see why a coach would go for it there (assuming you’re trying to win in regulation).
oops.
1 Like
I think this is, in general, a valid point. The Lions are hyper aggressive though. They proved against the Rams they aren’t afraid to go for the jugular when conventional wisdom says to run. They came out on that next drive passing when conventional wisdom says to run.
So while cutting the lead to 6 may affect how some offenses play, I think the Lions are going to be aggressive regardless.
Edited to add: in that situation you a reliant on your defense to get a stop with enough time on the clock to get a touchdown. Having an offense be more aggressive is actually likely beneficial, because there’s a greater chance for a stopped clock. So I’d much rather be down 6 and put the pressure on the other team.
1 Like
Olsen has the best explanation for all of it here. Gotta love him. I hate the MFer who hits on 18 at the blackjack table.
1 Like
It all depends on probabilities. If the probability of 2 points is less than half of the probability of 1 point, you’re right. If the probability of 2 is more than half of the probability of 1, going for 2 every time makes sense. With some exceptions of course. If you’re up by 16, going for 1 to make it a three score game (up 17) is preferable to going for 2 with a chance of leaving it at two score (up 16) game. In that case, goin for 2 to go up by 18 makes no sense, IMHO.
Of course, no one knows these probabilities for a particular event.
Don’t confuse independence with saying you can’t add the cumulative probability of multiple independent events. I would think that’s what @JimInTN is referring to as additivity.
In the bottom right under Olson, it says “commentator for Fox,” but he never mentions Fox once!?! I smell an ass whipping coming.
The players you have are more important than the analytics.
For example 4th and 1 on your own 40 yard line is basically a gimmie for the Eagles with the tush push with Hurts. But not for some other teams that struggle to run the ball or that has a weak center. Remember Raiola getting blown back for years
1 Like
NYTimes quotes 47.5% sine 2015. I would say you need to eliminate all 2 point attempts that happened from 1 yard out also because of penalty as @stephenboyd57 mentioned teams like Philly are excellent at that which most teams are not.
You multiply for cumulative independent events.
Not sure what you want to add (please provide example so I can see what you are saying).
Chance of 1 head on one flip =50%
Chance of 2 heads on two flips= 0.5^2 *100= 25%
Chance of 3 heads on three flips = 0.5^3 *100= 12.5%
etc.
But lets look at it a different way.
47.5% conversion rate for 2 points
so
.475*2= .95
94% for making xp (will make this 94 because at 95 it will come out same).
.94*1= .94
You are looking at .01 point difference. Will take a lot of games to make up that extra 1 point from pure score standpoint (100 TDs needed).
To be clear what TB did here made sense. This is purely message board discussion about going for 2 every time as a better play without looking at exceptions to rules as that is a long list.
1 Like
Greg Olsen says the outcome doesn’t matter doesn’t mean the decision is right or wrong?
In what world…
Newton Law was wrong then?
Doubt it
As he said, hitting on 18 in blackjack isn’t the right move, even if you’re lucky enough to get a 3. That’s bad process, even if they outcome was a good one.
Hitting on 16 when the dealer is showing a face card is good process, even if you bust (the outcome was bad). Over time the strategy will prove correct.
It’s the decision that matters. Make the right decisions over time and you will come out on top.
I’m not saying what you wrote is wrong
Heck I would probably even went for 2 like TB
But the analytics crowd fails to realize that you don’t get to make these decisions over and over and over.
If you wanna dictate the game sure go for it
If you’re confident in your team/play …game flow sure go for it
This is why I think MCDC gets it. He knows how to pick and choose his shots.
A lot of these coaches don’t
1 Like
I suspect that whenever we’re down 14, late in a game, and score a TD to go down 8 Campbell will go for 2 every time. Didn’t have that chance this year (against Chicago we were down 13 so going for 2 was unnecessary).
I actually think Dan is more aggressive than analytics tells him to be. Analytics tells him not to go for it on 4th and 2 against the Chargers or 4th and 4 against the packers last year. But he did. And analytics certainly didn’t advocate for going for 2 against Dallas after the penalties, but he did anyway.
It feels like he picks and chooses his shots because we’ve been conditioned for him to go for it. Like our first score against TB when he settled for a FG. We all thought that was him playing it conservative, but in reality that was an easy decision analytics-wise. Going for it on 4th against the Rams was the wrong decision according to analytics. He goes against the numbers a lot, but in pursuit of more aggression.
I agree that every decision requires taking into account more than just the analytics. But if anything I think it makes Dan more aggressive than the numbers tell him to be.
2 Likes
That’s the point I’m trying to make. He takes into a lot of other factors other than JUST the analytics. Which allows him to pick and choose
2 Likes