What on earth was that Spa grid about? It was the question everyone was asking in the paddock, and even the participants seemed to have as little idea as everyone else. Some of it was readily explainable, but by no means all.
There were the conspiracy theories, and you could see the logic. With F1 needing to ensure as many cars and teams as possible are around for next year, those three teams most in need of help - Force India (budget struggles), BMW (looking for a buyer) and Toyota (perceived as needing encouragement to stay) - were suddenly at the front.
Were the puppet masters pulling the strings behind the scenes, perhaps through the mechanism of the control tyres? Hard to see how it could be done, even if you believed it. How would you know in advance, at Spa of all places, what the temperatures were going to be to ensure the required performance levels?
If not that, then what?
The Force India VJM02 has been improving for most of the season and the Valencia upgrade was worth a huge 0.7s. From the start of the season to now, the team's simulation tools suggest the car has found over 1.3s, a bigger improvement than anyone else. Under the control of senior McLaren engineer Simon Roberts, it's a well-run little operation and the car's basic aero configuration is very sound. Probably better than McLaren's, ironically.
At Spa, the cars were running in low-downforce configuration for the first time all year and this has revealed a rather different order of performance compared to high-downforce trim. It seems that what downforce the Force India produces in low-wing form comes at less cost in drag than most. It also seems to hang onto a greater proportion of its downforce than its rivals.
The BMW seems to follow a similar performance profile, in its Valencia upgrade form at least. The other suddenly quick car, the Toyota, has an optimum downforce/drag trade-off at a point further up the downforce scale. It was super-quick in the high-downforce demands of sector two, making up for its deficit down the straights of sectors one and three.
But that's only part of it. It's hard to believe the Brawns and Red Bulls are suddenly aerodynamically poor in this low-downforce part of their aero maps. A clue as to what might be happening could be the Brawn's lack of lap time response to aerodynamic improvements made since mid-season.
"The windtunnel says they work and there's no question they've added downforce," says Ross Brawn, "but it wasn't translating through the tyres."
With continual car development but zero development of the control Bridgestones, it seems as if the tyres are effectively 'saturated'. No matter how much harder you press them down, they have no more to give. At the start of the year perhaps only the best cars had reached this saturation level. With the development programmes of the others, now they've apparently reached the same point.
Meanwhile, the continued development of the Brawns and Red Bulls is not reaping the expected rewards.
We have in effect engineered, by accident, a control formula. The performance difference between the fastest and slowest cars was already very small - just over 1s typically. By moving to a different part of the aero map, that small difference has had the effect of juggling the order.
"Yes, there's no question the tyre is now the main constraint," says Brawn. "It's the secret now to success in F1. It's a false limit and we're all struggling to understand that. I think that's what's given us the unusual grid and that's probably to the benefit of the sport."
Not so sure about the last bit. Do we really want something with the same effect as a control formula, while retaining the expense of bespoke engineering? Sounds like the worst of both worlds.
This would all disappear if we returned to a tyre war.
"But that would be inappropriately expensive," says Ross. "I think that stopped at just the right time." But what about a tyre war with no testing? "Oh," he says, with a big smile. "Then you'd have very speculative tyre development and there'd be some pretty dramatic swings."
Exactly. Sounds perfect to me.