Onside kick to increase number of offensive series?

Submitted by oldcityblue on October 6th, 2010 at 4:51 PM

We could get to see more Denard.

If that is not enough, please proceed.

Considering our TOP is probably going to be significantly less than our competition this year, and  knowing our defensive woes - some thoughts about keeping them off the field so our offense can continue to feast:

If our defense is going to be hard pressed to perform under normal circumstances, we shouldn't be surprised if they allow a lot of yards after being on the field 65% of the time.

Certainly, our offense will move quickly. However, instead of kicking off to watch another 12 play drive consisting of at least two 4th-and-whatever conversions because no one wants to give the ball back to Denard,  would it make sense to gamble on getting another immediate possession?

 Also, if we plan on being in shootouts the whole year, converting even one onside kick would significantly increase our chances of winning due to the extra possession.

Oregon was being outplayed by Stanford and they used the onside kick as a way to turn the tide, gain momentum and by keeping the Cardinal defense on the field, they subsequently took control. This play was key in  changing the outcome of the game, and was well executed. (planned strategy)  The same could be said about the onside kick for the Saints vs. Colts in Super Bowl XLIV.

In both instances, the move to the onside kick was used because they couldn't stop the other teams' offense. (sound familiar?)  

 Instead of using the onside kick as an expected last ditch effort to get the ball back with 2 minutes left,  how about using it as a planned play that the receiving team has to account for? We are going to be in a shootout anyway, why not gamble? If we fail and they field the kick cleanly, would it be that bad? (with the condition of our kicking game, it's no gimmie the other team won't  start on the 40 anyway.)

However if we convert the onside kick, much more fun will ensue. For Michigan 2010, more offensive series than the other team = win.

Get the ball back. Score again. Control the momentum.

 Also, we could get to see more Denard. 

- - - - - - - - - - -

Weird side note, if Dantonio is willing to use trickery...does any of this make sense for Sparty against us?



October 6th, 2010 at 6:06 PM ^

I have the same feeling.  I was sort  of surprised Indiana didn't try that - or even Bowling Green - since they couldn't stop us.  If you can't stop the other team, what difference does it make if you give them the ball on your own 40 as a result of a failed onside kick?  Mathlete?  Odds?  Statistical backup?


October 6th, 2010 at 5:04 PM ^

I'm not a kicking expert, so I'm not sure what the difficulty level of an onside kick is relative to other kicking skills. So, the recent kicking woes may or may not factor into that decision because you don't want to just hand the ball over to the opponent even if it feels like the opponent would get to that point on the field had you kicked it off anyways. However, it's not like they don't practice onside kicks, so I'm sure it's in the bag of tricks, maybe we can get Mathlete to calculate the potential point pay off or something.

Edit: I know onside kicks are dificult, just not sure how much skill/luck, etcetera is involved, and what that says about the odds of our kickers having a realistic chance to give the hands team a shot at recovering the kick.

Roger Mexico

October 6th, 2010 at 5:01 PM ^

Statistically, maybe it wouldn't hurt.  You'd have to compare the expected recovery rate of onside kicks to the chance of your defense getting a stop on a short field (as opposed to a longer field).


But, in a year where RR's job is on the line, etc., do you want him making strategic gambles this big, and being perceived as a 'rogue' coach (think Les Miles).  If it works, it might be worth it, but otherwise it's asking for trouble.


October 6th, 2010 at 5:10 PM ^

Surprise onside kicks are recovered at a better than 50% rate (I want to say around 60%).  Expected onside kicks are recovered much less frequently - like 20% of the time (though this percentage is inching upward as kickers get better at it).  I wouldn't mind trying a surprise at some point this season.

I don't think RR's job security (if it's even an issue at this point) should ever be a consideration in gameplanning.  You have to do what you can to win. 

Zone Left

October 6th, 2010 at 5:36 PM ^

Those numbers are about right.  I wouldn't be surprised (ha!) to see some trickery pulled out against State.  Michigan hasn't really done anything yet this year.

As for job security, redshirt decisions may or may not be affected by job security, but yeah, you do what you need to do to win.


October 6th, 2010 at 5:02 PM ^

The "bend don't break" defense doesn't work when bending allows your opponent into the endzone immediately.

With our defense our best chance is getting the other team into long drives, forcing them to dink and dunk their way down the field and make mistakes. Starting them out at our own 40 automatically doesn't help our case at all, especially with the small chances that an onside kick actually works.


October 6th, 2010 at 5:29 PM ^

1. Starting from your own 20 yard line, the offense needs to travel 50 yards to get in scoring range. 

2. Starting from the Mich 45 yard line, the offense needs to travel approximately 15 yards to get in scoring range.

Even if you average the NCAA median, approximately 5 yards per play, by kicking the ball off regularly would result in 10 successful plays needed to reach field goal range, while an unsuccessful onside kick would only require 3 successful plays.  So say the onside kick works 25% of the time once they expect you to do so, that means they have only 15 yards to get 3 points, 3 out of 4 times, or in other words, 9 points.  The one time you get the onside kick, say you score the touchdown, that's 7 points.  As you can see, we're already at a disadvantage.  And that is assuming we score the touchdown and they only kick field goals.


October 6th, 2010 at 5:52 PM ^

Unless this strategy is employed at kick-off.  I see what you're saying though, but the point I was trying to make is that they would only need 15 yards for the minimum point opportunity.  Our defense will give up 15 yards practically every time.  On the other hand, our defense will give up 50 yards only some of the time.  That's why I would feel safer kicking it off.


October 6th, 2010 at 6:04 PM ^

You can't assume that they score every time. There hasn't been a team this season that has scored on more than half their drives against our defense. Massachusetts scored on exactly half and one was a FG. If you give teams the ball in FG range, you are essentially giving them 4 opportunities and increasing their chances of scoring dramatically in exchanges for our 2 extra chances which we would statistically give us about 1 TD.


October 6th, 2010 at 6:00 PM ^

I think your 9 points  vs 7 points thought might not be entirely accurate because it doesn't account for the possessions that Michigan will already be getting after the other team converts a field goal/td after we attempt the onsides kick.

I would think that if we convert 1 out of 4 then wouldn't the points look like this:

THEM - 9 points (3 possessions)

US - 28 points (4 possessions)

I mean you could even say, that if they stop us once and we only get 3 touchdowns (21 points) on the four drives and they get 3 field goals and a touchdown (16 points), we are still statistically at an advantage, if our defense can bend but not break more times than not.

But I do agree with you that is assuming a lot.


EDIT: You can disregard this as it appears I wasn't the first to bring this to light... That's what I get for having a thought to post, but then having to actually do WORK at work and postponing the post.


October 6th, 2010 at 5:55 PM ^


In the Indiana game, they had 13 drives broken down like this:

5 TDs (11 plays, 13, 5, 12, 13)
4 Punts (3, 6, 5, 4)
2 Turnover on Downs (5, 12)
1 Interceptions (9)
1 Desperation drive at the end of the game (2)

Giving them the ball on the 40 means 3 punts are FG attempts or 4th conversion attempts. One TOD is a TD, the other could also be a FG.

Assuming they kick FGs early in the game, that gives them 6 TDs and 4 FG attempts. Take away 1 drive from them and give one to us and you are all of a sudden 2 points ahead after 13/11 drives with a quarter-ish of the game left.

Assuming they are aggressive and go for it on 4th down, given a 50% conversion rate, that's 8 TDs. Given the same scenario above and you are tied after 13/11 drives.

Our offense only scored on half their drives, so even if you expect to recover 2 onside kicks, statistically you aren't scoring again and they get the ball anyway.


October 6th, 2010 at 7:30 PM ^

Longer drives = more plays. Does more plays = more turnovers? Assuming our defense won't get them off the field by preventing first downs (which isn't entirely true), bend but don't break could create turnovers to get stop the drive. Turnovers aren't completely random - and Sparty did have three on Saturday. Chappel is a very good QB who didn't miss on too many throws - but even he threw a terrible pick. The "death by a million papercuts/bend don't break" strategy may prove more effective if we manage to create some turnovers.

That said a surprise onside kick would be something I'd like to see against Sparty.

Blue in Seattle

October 7th, 2010 at 1:11 PM ^

The key point you are missing is that Indiana completed that long drive with over 10 plays.  Each play is a statistical chance to be stopped.

That is why coaches of a weak or inexperienced defense apply tactics that focus on stopping the big yardage play over allowing better chance at short yardage plays.  Not because it offers more chances for the weak defense to stop, but because it offers more opportunities for the offense to make a mistake.

Indiana clearly is an experienced offense and last Saturday made few mistakes.  One interception the whole day from a lot of passing attempts.

What you need to factor in is not, "hey I saw two different teams complete one long drive in a game once" to "how many times does an opposing offense complete a long drive ( 80 yards being the typical expectation from a touchback) for a touchdown versus a short drive (say on the 50 or better, like after a failed onside kick) and then try to compare that to the opportunity percentage from the success of onside kicks.

Even subjectively I think you can rule out onside kicking all the time over kicking it long.

The biggest factor making onside kicks worth it statistically is when time is a factor (there will be no more opportunites to score no matter waht), OR if there will be a huge surprise factor to it, thus upping the success percentage.

So the Purdue onside kick was of the surprise variety (and there is nothing like success to prove the wisdom of risky decisions), and every onside kick with time expiring is basically low percent chance compared to zero percent chance.

I don't think you need to bring in math everytime to make a reasonable conclusion, or to shape the problem down to some good guessing, but just saying, "hey I saw it once" is a pretty weak argument.  Especially considering the wealth of data and previous conversation on this site that you could have tapped into.

Never Punt with Denard


October 6th, 2010 at 5:28 PM ^

Going forward, please do not mention last year again.  That team is an entirely different team that the 2010 team.  That's not to say that this team is without problems, but there is no reason to mention last year, especially a drive in a game by a team with Juice Williams behind center.


October 6th, 2010 at 5:27 PM ^

Starting them out at our own 40 automatically doesn't help our case at all, especially with the small chances that an onside kick actually works.

Don't confuse surprise onside kicks with expected ones.  The former succeed more often than they fail.


October 6th, 2010 at 5:03 PM ^

Robinson is fast enough that if he played on special teams that they could kick it to him.  He could catch it midair and go for the score immediately.  He can do it all!


October 6th, 2010 at 5:08 PM ^

I initially dismissed your idea as insane.

Then I thought about it. 

On the theory that our defense is going to give up 7 on just about every drive, why not risk it?  What is the worst that can happen?  State (or whatever team we play) will get good field position and score? Ok, so what?  They were going to score anyway.  But, as you said, if even 1 on-side kick results in us getting the ball back, then +7 for us.

Here are the problems with your plan, though:

1.  Our special teams / kicking / punt coverage has been not great.  They are also young.  Successfully working the onside kick is very difficult, even if performed by the best, most experienced personnel.  More often than not, this will just result on State (or whoever) getting the ball around the 50.

2.  We don't get scored on EVERY time - just most of the time.  So, while there is the chance of +7 in our favor, there is also the chance that Stete gets 7 (based on great field position)  on a drive where we might have stopped them.

3.  Element of surprise - if we do this too much, opposing teams will be on the lookout for it, reducing our success rate.

Here's how I would use your strategy:  If we are in a game where our offense is getting scored consistently (3 straight drives or 4 of 5), break it out.  At that point, what's the difference. 

Braylon 5 Hour…

October 6th, 2010 at 5:12 PM ^

What's next? take the defense off the field completely so that they score on the first play and we win the time of possession battle? Our defense needs to make enough stops to give our offense a chance to win. Thankfully, that is not a very high number.  There aren't any magic schemes or approaches left.  It's just going to come down to coaching and execution by the players to see if we can get any more out of our current talent.  If they can start to get more stops, we'll see more offense and probably more wins. Easy enough? 


October 6th, 2010 at 5:27 PM ^

I think that we have tried the whole "defense needs to make stops" "coaching and execution" thing, and we saw Indiana and UMass (UMass!!) move the ball on us with ease.  Yes, Chappell is a good QB, but Indiana is about as 1 dimensional of a team as there is.  And they pretty much just did the same thing all game - small dinks and dunks, short yardage passes, mind-numbing drives.

I am not advocating anything drastic, but if our defense is going to have ANY success, it is not going to be with basic "coaching and execution" because, quite frankly, we are outmatched in terms of talent, depth and experience.  We will need creativity to have success, and doing something like this once in a while to take a risk is not a bad idea.  And, I hope that this is not the only creative thing that our coaches do on defense.  I would expect that GERG, who has had decades of experience, will be able to come up with a few other creative wrinkles to help us out. 

Braylon 5 Hour…

October 6th, 2010 at 5:35 PM ^

Well I agree with what you are saying, my only point is we're not going to be able to just permanently keep our defense off the field by scoring a TD and then recovering onsides kicks over and over again.  To be a really successful team, the defense is going to need to do more.  I just don't think there's a magic scheme that suddenly makes Ezeh David Harris and JT Floyd into Leon Hall. So I think that's where coaching comes in, either these guys have to learn more to better maximize their abilities (and for the young guys to improve), or we just have to outscore everyone and make no mistakes on offense. 


October 6th, 2010 at 5:16 PM ^

This assumes our kicker would be able to properly execute an onside kick consistently.  I wouldn't do it unless we were huge underdogs to someone.


October 6th, 2010 at 5:17 PM ^

I think the problem with this idea, as it has been mentioned before, is that our defense relies on getting the opposing offense to make mistakes.  With a much shorter field, the opportunity to do so is much less.  Secondly, you're also assuming that our offense will score every time we touch the ball.  As prolific our offense is, that's not going to be the case either.  This is a very high risk/high reward scenario and I doubt it would be profitable over time.

Promote RichRod

October 6th, 2010 at 5:33 PM ^

First we'd either have to do it on the first kickoff for maximum surprise or (recommended) kick once or twice to see how the kick coverage is reacting.  If the front line is bailing early just dribble it straight ahead and grab it.

Obviously, this isn't some sort of sustainable strategy but I'd say the rewards offset the risks to give it a shot.  The risk is giving up 30 yards of field position--our kick coverage typically downs the returner around the 30 (weak kicks and meh coverage), whereas a failed onside will give it to the other team on your own 40.  So...what are these 30 yards worth?  IMO, not that much.  The bend don't break defense is the weakest through the middle of the field.  The idea is to shorten the field and then firm up the zone.  Obviously, we lose chances at a big play (turnover) through the middle of the field, but let's be honest...at this point any decent opposing offense is really only required not to drool on itself to march it from their 30 to our 40.  We still have the opportunity to force a 3 and out from the 40 or a long, difficult FG attempt.  If we happen to give up the TD it won't be the end of the world, either.

The upside, OTOH is tremendous.  It would provide a big momentum boost.  Our offense is awesome.  We could go up 2 scores instead of trading TDs.  We keep the defense off the field for another 90 seconds while we go 60 yards for the TD.

I was actually hoping for one of these against IU but it's probably saved for a better opponent.


October 6th, 2010 at 5:56 PM ^

I think that we could try this every once in a while, the surprise onside kick is a cool idea and can change the momentum quickly.  I think the more effective way to control the game would be to run the play clock down more often.  I LOVE that we have the ability to score quickly, but the better way to keep the defense off the field (especially with a lead) is to slow the game down a little.  Some of the best drives in Michigan history have been those long 8 minute drives that just make statements during the game, or salt the game away at the end.  Again scoring quickly is amazing, and I don't want us to become conservative, but maybe a little less of the Jet tempo and a little more of milking the full 40 of the play clock can be a very effective tool to help the defense.


October 6th, 2010 at 6:24 PM ^

the Les Miles MGoBlog member account!

Les (or shall I call you oldcityblue) your tendency for sane-sounding madness has exposed your true identity!


October 6th, 2010 at 6:46 PM ^

..that I think the acts of any coach, and I can't imagine that I've already said this, of the best team gives that team the distinct impression of putting themselves in the position where once victory is acheived, there is something gained. That's my issue, and the implications should be clear.


October 6th, 2010 at 6:34 PM ^

I only want to see an onside kick if Michigan is up 28+ points(which I would love to see, but do not expect).  I would love to see a 70 on that scoreboard this weekend.


October 6th, 2010 at 7:49 PM ^

I thought Chip Kelly's onside kick was a great call and it also worked out well for Purdue last year. Obviously it's high risk but I think in certain situations it's well worth it. In addition I think we need to be more aggresive on 4th downs. Within our 40 to Opp 20, I think we should be essentially playing for four downs from first down. Mathlete has shown good statistical support of this argument, but I think it's also just common sense. We simply can't afford to not score on drives.

Some have said shortening the field puts our bend but don't break defense at a big disadvantage. Well that might be true, except I'd say a more accurate description of our defense is bend and then break.


October 6th, 2010 at 11:58 PM ^

totally irrelevant, but on the playstation i never punt or kick field goals and ALWAYS onside.  The only reason why this seems plausible now is because on playstation I usually get denard 200 rushing and 300 passing or there abouts.  Life is starting to immitate art.


October 7th, 2010 at 12:21 AM ^

The Saints situayion was very clever, not only did we get an extra possesion, we took one away from Peyton Manning, a 14 point swing. The result? Saints win by 14.