Jump to content

Reich in large part may have lost due to his reliance on analytics...a cautionary tale


Big Turk

Recommended Posts

1 hour ago, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.

 

It's always a great idea until you fail.

  • Like (+1) 1
Link to comment
Share on other sites

1 hour ago, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.

That's not how it works.  It's not about finding the "right" answer in a specific game.


It is about doing what brings with it the highest probably of success.


But it's just a probability, which means there is a chance to be unsuccessful too.

 

If I flip a coin 10 times in a row, I am unlikely to get heads 10 times, but you might.

 

 

  • Like (+1) 2
Link to comment
Share on other sites

I agree with him going for it on 4th and goal. The way the Bills offense was playing at the time, it looked like the odds of them marching down the field was unlikely. We got bailed out by a slightly poor throw. As for going for 2, it evened out when they converted on the next TD.

  • Like (+1) 1
Link to comment
Share on other sites

1 hour ago, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.

Coaches have to go by feel and gut irrespective of analytics.  On the goal line, Rivers had been converting short yardage plays prior and their D was keeping our O pinned prior.  So I can't disagree with the decision.  On going for two...I probably would have kicked it to make it a 7 pt game.

Link to comment
Share on other sites

1 hour ago, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.

 

Going for two didn't really hurt them at all, as they missed that time, but made the next one so netted a total of two points.  If they had kicked the first, I'm about 99% certain, they'd have also kicked the second so same outcome, unless we again jumped Offside on the kick.

 

Going for it on 4th down, yeah didn't work out very well, though I'm guessing part of Reich's thinking may have been based on figuring the Bills were likely going to score 30 to 40 points in the game so he felt they needed more points too.

  • Like (+1) 1
Link to comment
Share on other sites

6 minutes ago, Ed_Formerly_of_Roch said:

 

Going for two didn't really hurt them at all, as they missed that time, but made the next one so netted a total of two points.  If they had kicked the first, I'm about 99% certain, they'd have also kicked the second so same outcome, unless we again jumped Offside on the kick.

 

Going for it on 4th down, yeah didn't work out very well, though I'm guessing part of Reich's thinking may have been based on figuring the Bills were likely going to score 30 to 40 points in the game so he felt they needed more points too.

IF we're playing KC for the crown, I'd want us to go for it on 4th down, and go for 2 from the 1.  When you know a team can score big, you need to take some risks.  If both work, they prolly win.  Frank can stand before the Indy team and fans and say he believed in his team with those decisions.  Go Bills!! 

  • Like (+1) 1
Link to comment
Share on other sites

2 hours ago, Big Turk said:

 

But it does happen...that is kinds the point...analytics says something likely won't happen but it did...

 

The Bills have done it several times this year.   Frank should have known. 

 

Link to comment
Share on other sites

3 hours ago, Process said:

I think both decisions were right. Can't be results oriented. They clearly were scared if the bills offense and wanted to be aggressive. I like it. 

 

The final two minutes though, that's another story. Specifically after the bs non fumble. They completely botched it.

 

That's probably more on Rivers though.


This is along the lines of what I think.  I think Reich is getting too much crap for the call.  You don’t beat scoring teams like the Bills with FG’s.  I think he saw an opportunity to go up 10 on the Bills and put a lot of pressure on them going into the second half.  The logic was good and even the play that was called was a good one.  The execution with Rivers and Pittman was just a little off.  So while the result was poor the thinking was good.

  • Like (+1) 2
Link to comment
Share on other sites

3 hours ago, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.


 

Analytics is a tool you are correct, but since you do not know when it is going to work and when it isn’t - you either have to be all in or not.  It is easy after the fact to second guess things, but the analytics should provide you with reasons.

 

I had no issue with them going for it on 4th down - I believe that decision was made once they got inside the 5, but the 3rd down play was terrible and really should have changed their mind to the FG to build their lead.  There was a chance for the Bills to score, but they should have taken the points when they lost yardage on 3rd down.

 

Going for 2 was a bit mind blowing to me because it had little impact if they got it.  It could have prevented a tie, but getting/failing on 2 pts. did not significantly increase their chances.  In the end it balanced out with the successful try, but then that was critical.

 

I thought Frank played an ultra conservative gameplan, but then made ultra aggressive decisions and that does not always go together.  If you are going to play a run heavy, short passing attack to keep the clock moving and really limit possession- then you need to score when you have chances and he cost the team points that mattered in the end.

Link to comment
Share on other sites

5 hours ago, Big Turk said:

 

But it does happen...that is kinds the point...analytics says something likely won't happen but it did...

 

It's very easy to point out when analytics doesn't work, but often goes unmentioned when it does work. If an event has 80/20 chance of success, it doesn't mean you made the wrong decision if the 20% outcome arises. I think that's the most difficult thing for fans to understand.

 

I also think that the decision to go for 2 had more to do with the stadium and their kicker. Blankenship was having an amazing season, but he missed 2 FG's last week, and another one today. Reich knows as well as anyone how challenging it is to kick here in January, especially for a rookie kicker. I'm surprised the broadcast crew didn't touch on this more.

 

In the end, the Bills were the slightly better team today. Our kicker made all his kicks, their's didn't.

Link to comment
Share on other sites

22 minutes ago, QCity said:

 

It's very easy to point out when analytics doesn't work, but often goes unmentioned when it does work. If an event has 80/20 chance of success, it doesn't mean you made the wrong decision if the 20% outcome arises. I think that's the most difficult thing for fans to understand.

 

I also think that the decision to go for 2 had more to do with the stadium and their kicker. Blankenship was having an amazing season, but he missed 2 FG's last week, and another one today. Reich knows as well as anyone how challenging it is to kick here in January, especially for a rookie kicker. I'm surprised the broadcast crew didn't touch on this more.

 

In the end, the Bills were the slightly better team today. Our kicker made all his kicks, their's didn't.

 

The thing I am impressed with most is the Colts played a near perfect game and brought their A game today. We played OK but we brought our B- game, especially on D and Daboll's playcalling and still won.

Link to comment
Share on other sites

The thing missing here is that broad analytics treat situations as though "everything is equal."

 

They do not take into consideration the emotion, momentum, or intangibles that play into a given situation.  If you're playing against the Jets, for instance, giving them the ball late in the 2nd quarter at their own 4 gives them nearly zero probability to score.  But Allen and the Bills?  Not so much, and it turned out that the Bills did just that, and stole the momentum from the Colts going into halftime.

 

FWIW, I liked the 2pt play after the penalty on their part, but I thought the 4th down call was too aggressive given the situation and what the turnover on downs meant.

Link to comment
Share on other sites

On 1/9/2021 at 6:27 PM, Bferra13 said:

Going for it 4th and goal from the 4 has been standard? Not the NFL I've been watching. 

 

The last 2-3 years it has. 

 

The same NFL that goes for it around the 40 yard line on 4th and 3. 

 

Game changed. Doug Peterson was the starter of this and his OC was Reich.

Link to comment
Share on other sites

1 minute ago, BillsToast said:

 

The last 2-3 years it has. 

 

The same NFL that goes for it around the 40 yard line on 4th and 3. 

 

Game changed. Doug Peterson was the starter of this and his OC was Reich.

Well its doubly dumb in the playoffs at that moment with those circumstances. 

Link to comment
Share on other sites

On 1/9/2021 at 7:04 PM, Big Turk said:

Reich credited his decision to go for it on 4th and goal from the 4 late in the 1st half and his decision to go for 2 that he didn't get from the 1 yard line after the Bills penalty to analytics...

 

This is why following analytics is such a tricky thing...small sample size. Yes, out of 100 times, you might be better off with analytics 65 of them but those other 35 are not going to work and in 1 or 2 situations a game, there aren't enough opportunities for analytics to make up the difference.

 

Analytics should be a tool but buyer beware on relying on it to win amy single game. Over the course of 10 years it might win you more games than it loses you but it also might get you fired in year 3 if enough of them don't work in your favor.

 

On 1/9/2021 at 8:16 PM, PromoTheRobot said:

 

It's always a great idea until you fail.

These arguments are the reasons that professional football is still in the Stone Age. 

Link to comment
Share on other sites

36 minutes ago, Bferra13 said:

Well its doubly dumb in the playoffs at that moment with those circumstances. 

 

Not really. Where's the proof it is? 1 time is an anecdote. 

 

Pederson and Reich won a Superbowl with an inferior team using odds in their favor with a backup QB. 

Edited by BillsToast
Link to comment
Share on other sites

×
×
  • Create New...