New kickoff rule
Might be spot on!
Or not....but either way, I like the proposal.
Edj Analytics might be able to tell us the answer. The Eagles use it and made some critical Super Bowl decisions based off of the software.
Feel like the new kickoff rule would change some of those analytics. Because some of those numbers would include times an opponent was pinned inside the 25. But if every possession after a scoring play starts at the 25? That would change things.
I'm bored and was interested in this. I'm using the home team model for Expected Point Value based on Big Ten conference games from 2013-2014 found here: http://blog.minitab.com/blog/the-statistics-game/big-ten-4th-down-calcu…
The Expected Point Value of 1st & 10 from the 25 is 1.026.
The Expected Point Value of 1st & 10 from the 50 is 2.902.
So from the kicking teams perspective:
-1.026 = 2.902p - 2.902(1-p)
-1.026 = 2.902p -2.902 + 2.902p
1.876 = 5.804p
p = 1.876/5.804 =32%
The breakeven point is recovering 32% of their onside kicks.
The recover rate for non surprise onside kicks is roughly 20%, while it is 60% for surprise onside kicks.
It is statistically worth it to try for surprise onside kicks, but that has always been the case even before this new rule.
Looking at Michigan last year, they had 32 touchbacks and 37 returns for an average of 15 yards. Being as favorable as possible and assuming the return kicks result in an average field position of 1st & 10 from the 15:
32 X 25 = 800
37 X 15 = 555
1,355/69=19.6
The Expected Point Value of 1st & 10 from the 19 is .5758.
1.026 - .5758 = .4502.
If every return last year now results in a fair catch, Michigan loses 6 yards per kickoff. Their Expected Point Value is -.4502 less per kickoff than last year. Michigan averaged 5 kickoffs a game last year, which would be an Expected Point Value decrease of 2.251 per game. That said, teams are still going to try and return balls, and if your kickoff coverage is elite, you will still gain an advantage here. This is basically the worst case scenario for Michigan and it still only results in a 2.251 decrease in Expected Point Value per game for last season. That's not enough to really change your kickoff strategy dramatically.
Looking at the breakeven point last season for Michigan:
-.5758 = 2.902p - 2.902(1-p)
-.5758 = 2.902p -2.902 + 2.902p
2.3262 = 5.804p
p = 2.3262/5.804 = 40%
So this rule change just means Michigan would have to recover 8% less of their onside kicks to reach the breakeven point, but doesn't change the overall kickoff philosophy. It made sense for the them to try some surprise onside kicks last year, and it still makes sense for them to try surprise onside kicks this year.
TLDR: This rule change does not move the needle from it being worth it to try a surprise onside kick this year vs not being worth it last year. It has always been statistically worth it, it will be even more valuable to do it this year, but you're talking about a difference in Expected Point Value per kickoff of less than 0.5.
It is unlikely this rule change will really have any impact in how teams approach kickoffs.
wow...you must've been really bored
First off, great research. But I have a question: what’s the average starting position for a power 5 team on a kickoff? You say teams are still going to try and return kicks, but I don’t think that’s true. If you’re guaranteed to start at the 25, why try and return it if you’re going to end up at the 20? Or 22? It’s going to be the same as punts: don’t try and return it if you’re inside the 10. Except that’ll be like the 10 or 15 yard line for kickoffs now.
And an increase in surprise kicks is exactly what I’m talking about. If the new norm is going to be lots of fair catches and bored and not fully locked in coverage units I think you’ll see more surprise kicks. Now those may not happen at the 60% rate they’ve done in the past, but even if they drop to around 40%, if there’s now a 40% chance I can get the ball my own 50 in exchange for a 60% chance of giving up just 25 yards? I’m taking that gamble. Maybe not every game, but I think it’s going to be something you see more and more of.
April 13th, 2018 at 10:17 PM ^
Teams returned balls that were caught in the end zone, even against Michigan, so they will absolutely still return some balls that are caught at the 5.
As for why, because of the upside. If you fair catch it you'll start at the 25. If you return it, you may get a TD or have a 50 yard return. If you have an elite return man, hes gonna average 25+ yards a return, so on average you will start with better field position than if he fair catches everything. There really isn't that much downside to returning a ball caught at the 2 if you have a good return team or are playing against a bad kickoff coverage team. You're risking maybe 5 yards of field position for the chance to gain a helluva lot more. There will also be situations at the end of half's where teams won't have the time to drive the field and will be more likely to take a chance returning the ball as a result.
Its just like teams in college basketball that slow the pace or shoot a bunch of 3s because they are outmatched. There are different talent levels amongst teams, and kickoff returns give teams a chance to break the game open and crate a double digit point swing. Look at PSU-OSU last year. OSU absolutely dominated that game but only won by one. PSU had one kickoff return for a TD and another 59 yard return that set up a 23 yard TD drive. If they fair catch both they lose that game by 10+, instead they almost pulled it off despite being dominated.
If this new rule has the impact you are suggesting, and everybody just fair catches everything, that would pretty much eliminate the surprise onside kick. Teams would just have their hands team lined up all over the place and the players would never look to block and would just anticipate an onside kick. The surprise kick works because players are mentally getting ready to block somebody on the coverage team. If there are no more returns, they won't have to worry about blocking and will just focus exclusively on the ball.
I couldn't find any stats about the average starting field position on kickoff returns.
Keep this up and you will be the #1 draft pick for my fantasy statistics league.
Nice statistics, but also enjoyed your logic. Will we see your articles in the Economist this year or you waiting until 2019?
April 13th, 2018 at 11:25 PM ^
A couple of comments. I enjoy these discussions.
First, with "surprise" onside kicks, you're always dealing with a low sample size. And in this case, when the sample gets larger, percentage of recovery will drop significantly. I'd also imagine that the new rule will cause teams to be a little more prepared for onside kicks.
Second, per-possession analysis is tough because there is significant variation in number of possessions per game. Shorter fields increase possessions totals and end up putting a little more pressure on your own defense. If you can pull it off, that 13-play, 80-yard touchdown drive is worth quite a bit more than it seems.
So I'm not comfortable with addressing this question with an answer based on this low a percentage change (and you agree as well).
I'd probably conclude that it's not a horrible thing to work on onside kickoffs and try a surprise one every few games just to make sure opponents use time preparing for it. But I suspect unless there's something specific the coaches see in an opponent's kick return, it's not a good idea to try one in every game.
Delete
25 yards might not sound like a lot, but that's not the case here. Getting the ball at midfield is a huge plus for any offense.
The entire playbook is open and having such great field position you might be putting your defense on the hook for 4 downs versus 3. As teams may elect to go for it on 4th rather than punt or attempt a long field goal if they are able to move the ball even a little bit.
I agree - and onside kicks hardly ever ever ever work out. Bad strategy OP, regardless of the new rule!
Surprise on-side kicks are recovered far more often than when the receiving team expects one.
it's they a fact it a feeling that surprises are recovered a lot? wouldn't on side kicks no longer be a surprise in this scenario? I think it won't cause more on side kicks. i think we still pop it up inside the 10 and see what happens. just cause they can fair catch doesn't mean they will.
I think I read that surprise onsides kicks are recovered by the kicking team 60% of the time. But expected onsides are only recovered like 10-15% of the time.
Under this new rule, teams would be stupid to try to return a short pop up kick from inside the 10.
Teams aren’t going to return it if it’s popped up inside the 10. If you’re guaranteed starting at the 25, it’s going to be the same old adage they use for punts: don’t try and return it if it’s inside the 10. Let it go and get the field position. Except it’ll be and even stronger argument because there’s a 0% chance of them pinning you deep like they can on a punt if you let it go.
Because instead of needing 4 or more first downs (if you can limit the big play), a team would only need 2 first downs or one chunk play to have a decent field goal attempt. That’s probably 10 or so more plays that an offense has to properly execute without turning the ball over to our points on the board. Seems that onside kick, rarely recovered, would be a statistically bad choice for a team with a good, or even decent, defense.
the rule, but why do you think this would make teams try for onsides kicks? That's like going for it on 4th down every time because of the extra play. I could see fair catches being used on pooch kick offs.
Feels like if every kickoff is going to be a fair catch, teams will get lax with their focus which will result in more onside attempts to catch people sleeping. Plus, it’s now less of a gamble due to assured field position.
They won't sleep if a team shows a tendency to try an onside even once a game on average, let alone a few times a game.
It looked to me like most teams were contect with touchbacks when kicking off. We were an exception because James Foug was exceptionally talented at placing his kicks. We might have reverted back to kicking into the endzone with him graduating. Now we will for sure.
receiving has assured field position, not the team kicking off. I think you are a little confused.
I get that this rule is for safety . Am i being selfish in saying so what? I think this is a dumb rule . Might as well work to kick it throught the end zone each time .
This will be good for teams with great offenses .
One more reason that I cant wait for College Football season .
Rather than onside kicks, I can see an increase in squib/line drive kickoffs that hit at higher velocity and lower trajectory, potentially making it a more difficult ball to control, while also eliminating the possibility of a fair catch.
Exactly. It’s about to be squib city out there
that's what I was thinking. squib it, and see if you can get someone to mishandle it, panic, fumble, etc.
Wonder where the average squib or line drive kick gets returned to. Feels like right around the 25 or 30. And since you’d be guaranteeing a return of some sort, you’d be exposing yourself to more lost field position.
Same thing I was thinking. Why kick it high & deep? Hell, with this a kick out of bounds is only a 15 yard penalty. Squib it or onside every time. Let's see your guys who don't normally handle the ball figure out how to grab it off a bounce.
Teams will adjust. If you onside it regularly they'll just always have hands team guys at the front line.
Surprise onside kicks are worth it because they are recovered at such a high rate. Non surprise onside kicks are not worth it because they aren't recovered at a high enough rate.
that's why squib makes more sense than onside. kick it over the front line and make the deep man field a line drive/rolling ball off the ground.
all these theories seem easily disproven by the fact that most teams were willing to lick into the end zone regularly. why not try squibs and on sides instead of kicking into the end zone? I foresee no change.
because with the old rules, to force a decision, your squib would have to cross the goal line. and, the receiver could simply fall on it for a touchback.
now, you can squib it inside the 25 (much easier), and if the receiver simply falls on it, they dont get the ball advanced like a touchback. you are forcing them to return it, or accept worse field position.
You do realize this strategy could be used in the past- squib it to the 10.
I would love to see more squibs. I think probably makes the play more exciting and also less dangerous. Reduces the speed at which the receiving team is moving forward.
You're right this improves the math on onside kicks.
More than that I think we'll see more high and short kicks daring the up backs to field the ball around the 20-30 where a muff is likely to be a turnover. Urban used to do that at BGSU under the old rules; to great effect in a game vs. Toledo I attended way back in the day.
The very good "Alaskan Assassin" WR on BGSU at the time (Cole Magner for early 2000s MAC reverse pass enthusiasts) used to drive around campus in an Escalade FWIW....
This is dumb.
This is stupid.
This is moranic
This is asinine.
this is idiotic
this is genius
this is us is dumb.
This is ME
This is silly.
This is brilliant!
(in bizarro land)
You may see a small uptick in surprise inside kicks, but that's about it.
And the better your defense is, the less sense it makes to try for an onside kick.
I try to avoid surprise inside kicks at all costs.
I just had a dental implant put in today, and you made me laugh; which caused a stitch to break. Laughing and bleeding....it's what I do.
So sorry, sounds painful...of course, if that’s what you do, carry on.
As we routinely were placing kickoffs inside the 5 with good hang time and pinning teams inside the 20 to start their possessions.