The Irish used a relentless offensive game plan and a calculated defensive game plan to hold off a talented rival in a marquee matchup. SP+ didn’t view that performance so kindly, largely because patient offense and bend-but-don’t-break defense tend to work counter to the metrics that SP+ considers important. Could the Notre Dame offense have been more explosive? Maybe. Could the defense have been more aggressive and forced more Trojan errors? Possibly. At the end of the day, though, Chip Long’s George Foreman act and Clark Lea’s Floyd Mayweather act worked perfectly together and the Irish came away with a hard-fought win. Hard to argue with that. It’s also worth noting that Brian Kelly has now won three straight against USC and is 7-3 overall against them, which is the best mark for an Irish head coach since Lou Holtz closed out a 9-1-1 career against them.
Recapping the methodology again: These aren’t actual SP+ win probabilities, as SP+ creator Bill Connelly doesn’t publish those regularly after his preseason previews. We used a slightly tweaked version of the formula that the Reddit guy reverse-engineered to get close to published probabilities. That formula does some probability magic on the differential in SP+ rating for the two teams (with a bump for home field), and voilà .
Post-Week 7 Update
As you move down each column, you find the probability of Notre Dame owning that many wins at that point of the schedule. In the first row, you can see that the probability of owning 0 wins through week 1 is 0% and one win is 100%, since we already won. Same for the New Mexico row, and so on. In the last row, we’re tracking how much the probability of each win total has changed from the previous week. This is largely a function of our own outcomes, the quality of play that led to those outcomes, and SP+’s changing perceptions of our opponents.
The 8.9% figure at the end of the Stanford row reflects our probability of having 11 wins through 12 games – in other words, that’s our probability of winning out. That dropped a lot from last week based on Notre Dame’s own SP+ rating dropping rather precipitously. 9-3 is back to SP+’s most likely outcome, at 35.1%; very interesting, as I think at this point anything less than 11-1 would disappoint a very large portion of the fan base.
Week by Week Game Trends
Other Nuggets
- Tackling the “Big Three”: It felt a little silly to redefine the “Big Three” again with the win over USC in the books – I mean, do any of you consider Duke a big game? How about Navy? So I’ll just track Michigan and Stanford at this point. Not surprisingly, given the sharp drop in the win probability against Michigan, some kind of split with Michigan and Stanford is the current most likely outcome, at 56.1%. A sweep is next at 26.3%, while endless sadness trails at 17.5%.
- Taking care of business: Notre Dame’s expected win total dipped down again from 9.33 wins to 9.15 wins. If the Irish win all games in which SP+ heavily favors them (70% or higher win probability) that expectation scooches up to 10.03 wins, which is down from 10.16 last week. Michigan and Duke are the only non-heavy-Irish-favorite games remaining.
- Movers and shakers:Â Highlighting just the notable movers…
- I just don’t get the Michigan move, a whopping 13.5 point drop in win probability. They looked pretty bad for sizable stretches against a below-average Illinois team. But, again, they’ve figured out how to game the SP+ system, plus this year the conference adjustments are very friendly to Big Ten teams. Let’s see what happens this week when they travel to Happy Valley.
- I’m not going to react to too much else here, because I think a lot of it comes from a strategic game plan against USC that doesn’t play to SP+’s preferences, as noted above. The one other thing I would highlight, though, is the Navy win probability dropping 9.4 points after they dismantled Tulsa 45-17. I’m not sold on them yet, but there’s a much better chance today than there was preseason that they might provide some resistance.
I don’t have a problem with how SP+ judged our game. It was one with a specific game plan. But a game plan that BK himself basically said was a losing game plan once it was figured out. So the coaches made a strategic decision to use something that if USC figured out too quickly we’d be in trouble. They judged correctly that we could survive it and win, but one has to ask if that points to our team being still not quite good enough such that the coaches feel like their best plan is one that has some higher risk. All that SP+ is doing is saying those types of risks – if you make a living on them – will eventually catch you.
The really funny thing is that usually, I feel like our staff does a great job of adjusting at halftime, but we often start slow. In this one, it was almost like that fast start put the staff into cruise control. I know you’re coaching against Clay Helton, but you still have to expect him to make 1 or 2 halftime adjustments, right?
That being said, a win is a win. If the tackling is better against Michigan, I don’t think the gameplan will matter in the least, as I’m still not convinced their offense is anything worth worrying about.
The halftime score was 17-3, but just as importantly, it was 17 unanswered points in the second quarter. The offense was rolling, and the defensive bend-don’t-break strategy was working beautifully. USC hadn’t done anything well enough to require significant adjustments.
Sure, you expect USC to adjust, but it’s hard to ask your players to change their effective strategy in anticipation of unknowable shifts in the other locker room.
Agreed about UM. They don’t look great on offense or defense. I’m not predicting a comfortable victory, but I haven’t seen anything scary from Harbaugh’s 2019 Skunkbears.
The lack of love from the advanced statistics is disappointing but probably reflects how I feel. If this team can find another gear (a consistent one at that) that includes either an efficient or explosive passing game this is a top 10 team that goes 11-1, but I still have questions on that happening.
Maybe the bye is coming at the right time, get Armstrong and Crawford healthy and find an additional receiving option and start peaking in the second half
I’m not as high on this team as a lot of the commentariat, but, man, even I don’t think we’re more than twice as likely to go 8-4 than 11-1 with our remaining schedule. That seems really odd. I do think this team could have gone 8-4 or 7-5 against some previous ND schedules (e.g., if somehow we could time travel, I bet this team would be clear underdogs to 2011 ND), but I’m not seeing it this year.
Well 11-1 requires no trip-ups, no “bad games,” and no series of injuries of just the “wrong” players, (and also proving SP+ wrong about who’s better us or Mich.). Do I think 8-4 is highly unlikely? Yes. But I also don’t think this team is quite good enough to be the best team on the field 6 more times by a large enough margin that if anything “unlucky” happens we can overcome it. So I agree that 25% is a bit high for us winning 8 or fewer games, but if I was forced to bet 8-4 or 11-1… I probably take 8-4.
Interesting. I’d definitely go 11-1 over 8-4 at this point. I don’t see 3 more teams that can beat ND, and they haven’t lost a “trip up game” against a non-top 25 ranked team since, what 2016 when all hell broke lose? I think they’re a mature enough, strong team that can be trusted at this point to beat all the teams they should beat relative to losing 2+ of those games quickly in the next six games to start dropping them all. IMO, if they would have been on the 8-4 path that would have required losing to UVA or USC.
If you told me we’re going 11-1 or 8-4 for sure I’d be all over 11-1. The chances of us losing to Michigan and two of the (outside of Navy, I guess) very mediocre to bad teams we play in November strike me as being negligible.
8-4 requires Book getting hurt, IMO. But Jurk has looked sufficiently bad every time we’ve seen him that 8-4 seems very possible with him as the QB.
That statement is flatly untrue. He looked really bad in the spring game, for sure, but he’s been incredible in game action this year. He’s much more polished as the year has gone along and in his mop-up time been highly effective.
Incredible ?
It just happens that what we’re doing this year doesn’t agree with SP+ as much as it did in 2017 and 2018 – other advanced stats models are kinder to us. ESPN’s FPI, for example, gives us a 62% win probability against Michigan while 538 gives us 56%. Different strokes for different folks.
Does FPI, FEI, SP+, or 538 publish their predictive success rate? Against the spread vs straight up?
Bill Connelly tweets out how SP+ does against the spread, not sure about the others. This year SP+ started out like a house afire against the spread but has cooled off dramatically over the last couple of weeks. I believe that historically SP+ straight up is somewhere around 60%-65% and ATS is around 55%, not sure though.
He tweets out updates, I believe his last one was 57% ATS and 53% on O/U. I’m pretty sure these would be his best numbers ever. I’m in Vegas next weekend and I’ll definitely be looking for gaps in spreads and what SP+ has
i need some of that last stand of charlie weis
it’s like crack
ND was off 10 years ago this week, so the Last Stand was too. Don’t worry, it’ll be back for the Weis Death March. 🙂
More #sportsmath fun… I put together a second spreadsheet to track FPI win probabilities and will probably switch this column next week to cover both. Note that I wouldn’t consider either model necessarily superior to the other; they use very different methodologies and it’s just interesting to set the two side-by-side.
FPI, for this year at least (not sure about previous years), is way more impressed by ND (or perhaps unimpressed by its opponents, or both) than SP+. As noted earlier in the thread, FPI gives us a 62% chance of beating Michigan vs. the 36% chance SP+ gives us. Overall, FPI gives us a 32% chance of winning out and a 75% chance of being 10-2 or better, vs. 9% and 39% respectively from SP+.
Here are SP+/FPI win probabilities for all remaining games:
Michigan – 36% / 62%
VT – 82% / 95%
Duke – 67% / 78%
Navy – 74% / 93%
BC – 84% / 96%
Stanford – 72% / 78%
They differ very substantially across the board. It’ll be interesting to track this in the coming weeks, I think.
Another interesting #sportsmath finding is on 538’s college football predictor. According to their statistics, we are one of four top 25 teams with a less than 50% chance of making the playoffs even if we win out (they have our chances as 48%). The others are Texas (35%), Washington (9%), and Boise St (3%), two of which already have two losses. Notable conference-affiliated teams who have a higher chance than us of making the playoffs by winning out include Iowa St (60%), Arizona St (63%), and Wake Forest (78%).
538 also gives us a 56% chance to beat Michigan, and a 28% chance to win out. You can manipulate statistics by looking at what happens if each team wins their next one, loses their next one, or wins out. Our chances of playoffs go all the way up to 80% if we win out, and OU and LSU lose their next games, but 538 would like me to know that this combination occurs in less than 0.25% of simulations, and I should pick a more likely outcome to explore.
Of course, 538 also said that Hillary had a 74% chance to win the election, so…..
I do think that if we win out, we aren’t making it in. The committee loves wins over big time opponents, and they don’t seem to really care who you lose to. If Michigan ends up being our best win, I wouldn’t be shocked to see them end the season with 4 or 5 regular season losses. It would take a decent amount of chaos to get us in; not impossible, but it needs to start sooner rather than later.
“It would take a decent amount of chaos to get us in; not impossible, but it needs to start sooner rather than later.”
Disagree here on the “sooner rather than later” part. Consolidate the chaos as much as possible into the last 25% of the season. Gotta play into the recency bias and use the argument that ND is a totally different team than they were in September.
I think strength of schedule (including potential conference title games when considering winning out) is factored into the low probability of our making it. In this circumstance, not being in a conference hurts our chances of making the playoff, as our marquee games are really not as good as predicted. We don’t even have a team on our schedule that is outperforming preseason expectations, to balance out against the blue bloods not being great. Could be at the end of the year that wins over 9-3 UVa or Duke or even Louisville won’t look too bad, but all of those teams could just as likely be 7-5. Add that in to losing our only game against a top 10 team, and USC Michigan and Stanford greatly underperforming preseason expectations, and we just don’t have the resume to deserve it. That’s fine with me. I’ll take a win in the Cotton or Orange, no matter who the opponent is (currently I’m rooting for SMU)
Completely agree; there are some people who will say that missing the playoff would be a failed season, but I will 100% be ecstatic with an NY6 Bowl Win.
Cotton Vs SMU would be REALLY interesting. ND has a huge following in Dallas and draws well here every time we play. SMU is marketing itself as Dallas’ college team. I can imagine a tremendous atmosphere in Jerry World if that game happens
Let’s be real here: the best possible realistic scenario for ND this year is that we win out, then play Boise State, beat them by 2 TDs plus, and then get to complain forever that we would have won the title (secretly knowing full well that we would not have won the title).
Care to divulge who on this site you think would make that claim. Pretty realistic group here I’d say. I don’t think anyone thinks we’ll make the playoff nor is there anyone that thinks we’d beat the top 2-3 teams in the country. Most here, realize we’re going to have our hands full next week.
Sorry, I didn’t mean on the board here. I meant IRL performance complaining to the non-ND fans in your life.
Yeah, sign me up here (aside from the complaining, lol). I do think we’re a good team, but I don’t think we’re an elite team and I don’t think there’s any good reason to believe we’d see anything other than a repeat of last year’s Cotton Bowl if we got into the playoff. Plus pining for the fjords, I mean playoffs, too much takes our attention away from the fact that this has the potential to be a really good and fun year in and of itself.
Forget the playoffs. If it happens, great, but either way just enjoy the ride.
I agree…I don’t think even at 11-1 that ND would make it into the playoffs. (Though I think it’s entirely possible, with lots of carnage available in the B1G East and SEC West which would drive up ND’s ranking.)