Jump to content

Jets: only 22% chance of winning


stoicsentry
 Share

Recommended Posts

just thought you should all take note, accuscore was 1-3 last weekend. Keep it up ESPN, we love ya!

Screw Mike Golic, Stink, and all of the "experts". Schefter was 1 of 2 to get our game right. Count The Jets out this week, and next, and the next after that.

Rex has got something for the Pats, I know it!

Link to comment
Share on other sites

just thought you should all take note, accuscore was 1-3 last weekend.

I don't think you really understand how this works. To say that AccuScore is right or wrong in hindsight you have to round a probabilistic projection (Jets have a 22% chance of winning) into a binary one (Jets will lose). Evaluating the results of the model this way treats a 99%/1% projection exactly the same as a 51%/49%, ignoring that the whole point of the exercise is that not all outcomes are equally likely. 3.5:1 really isn't all that bad, and you have to remember that the very same model that put us at 22% to win before the game started would have projected us as solid favorites if you re-ran the sims after the first play.

Link to comment
Share on other sites

I don't think you really understand how this works. To say that AccuScore is right or wrong in hindsight you have to round a probabilistic projection (Jets have a 22% chance of winning) into a binary one (Jets will lose). Evaluating the results of the model this way treats a 99%/1% projection exactly the same as a 51%/49%, ignoring that the whole point of the exercise is that not all outcomes are equally likely. 3.5:1 really isn't all that bad, and you have to remember that the very same model that put us at 22% to win before the game started would have projected us as solid favorites if you re-ran the sims after the first play.

Placing any importance on a statistical model that comes from a pro-PATS sports network like ESPN is ridiculous. I like your anaylsis, but I will take it one step further, the projection does not include probability statistics nor does it include what we call in the engineering field a durability model. What that means in laymens terms is adding a fatigue factor to then forecast when something will "break". In this case it would be the defense vs. the offense of both teams using the 16 games as your baseline number. Then step two would be to take the defensive yards allowed by each team as the numerator. I know bear with me....once you have determined that number you then divide that number by the offensive yards and plays per game. That number then is compared to a "hidden value which would equate to an unknown value which is equal to an intangible affect".

This is where the 22% projected probability is useless as described by ESPN's accuscore. Because it only factors known quantities. But to be a true forecast you must add ALL factors to have a true probablity.

In my opinion since both teams split the games the intangible affect is much greater, which means this factor alone has more weight to it than a pure numerical number (i.e. number of TD's in the redzone vs. number of offensive plays required to get into the redzone). Which means to me that the Jets win this game as they have basically done all year, the score in the end is not indicative of actual field play.

Link to comment
Share on other sites

Because it only factors known quantities. But to be a true forecast you must add ALL factors to have a true probablity.

There's no such thing as the 'true' forecast you describe because the probability that any given event will occur at some point in the future depends on infinite variables which are constantly changing. At best they're reasonably ascertainable and changing in a predictable fashion, e.g., the probability that the sun will rise tomorrow is now .999etc. and will continue to approach 1 right up the the point it happens because the correlation coefficients on the variables that could actually cause it to move in the other direction, like a black hole or whatever, are infinitessimal. With football they're huge by comparison. Known quantities simply don't exist prospectively, and the idea that a projection has to account for the butterfly effect is insane.

Link to comment
Share on other sites

I don't think you really understand how this works. To say that AccuScore is right or wrong in hindsight you have to round a probabilistic projection (Jets have a 22% chance of winning) into a binary one (Jets will lose). Evaluating the results of the model this way treats a 99%/1% projection exactly the same as a 51%/49%, ignoring that the whole point of the exercise is that not all outcomes are equally likely. 3.5:1 really isn't all that bad, and you have to remember that the very same model that put us at 22% to win before the game started would have projected us as solid favorites if you re-ran the sims after the first play.

methinks you're wasting your time

Link to comment
Share on other sites

Placing any importance on a statistical model that comes from a pro-PATS sports network like ESPN is ridiculous. I like your anaylsis, but I will take it one step further, the projection does not include probability statistics nor does it include what we call in the engineering field a durability model. What that means in laymens terms is adding a fatigue factor to then forecast when something will "break". In this case it would be the defense vs. the offense of both teams using the 16 games as your baseline number. Then step two would be to take the defensive yards allowed by each team as the numerator. I know bear with me....once you have determined that number you then divide that number by the offensive yards and plays per game. That number then is compared to a "hidden value which would equate to an unknown value which is equal to an intangible affect".

This is where the 22% projected probability is useless as described by ESPN's accuscore. Because it only factors known quantities. But to be a true forecast you must add ALL factors to have a true probablity.

In my opinion since both teams split the games the intangible affect is much greater, which means this factor alone has more weight to it than a pure numerical number (i.e. number of TD's in the redzone vs. number of offensive plays required to get into the redzone). Which means to me that the Jets win this game as they have basically done all year, the score in the end is not indicative of actual field play.

Jets 28-Pat 21.....The score was not indicative of the game....We KICKED THEIR A$$E$ as I explained above.....the very same thing will happen this soming sunday at Pitt....

Link to comment
Share on other sites

There's no such thing as the 'true' forecast you describe because the probability that any given event will occur at some point in the future depends on infinite variables which are constantly changing. At best they're reasonably ascertainable and changing in a predictable fashion, e.g., the probability that the sun will rise tomorrow is now .999etc. and will continue to approach 1 right up the the point it happens because the correlation coefficients on the variables that could actually cause it to move in the other direction, like a black hole or whatever, are infinitessimal. With football they're huge by comparison. Known quantities simply don't exist prospectively, and the idea that a projection has to account for the butterfly effect is insane.

AS I Already said....I was right....and I was right about Pitt in the first matchup and will be right again this coming sunday.....

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...