Protour, analyses.
 

Race one.

It appears that the first race ended in 2 large pelotons. One group consisting of 40 riders, the other 110 riders.
They finished 3,26 Minutes apart.

Notable weird things.

-Reno's Rumbling Roadkills Gorazd Delalande wins the sprint, though he's the ONLY rider of that team in the group of 40!
-Other teams, in particular Misfits and Supersonics' Cyclingsquads had the vast part of their teams in the group of 40, yet somehow they did not manage to get a victory. This is possible but as they did not even finish top 7 something is probably not right there!
-The winner. Well, he is not a good rider at all! He has 4 climb, 5 descend, 2 time, 4 steer and 3 sprint. He also has just 2 experience. I will not believe that this dude is even close to the best sprinters in the group of 40.
-Another strange team is Rambobank. They finish 3rd, 4th and 5th. Yet, they are the only riders of this team in the group. It would be severely illogical if all three join up in the mass sprint! Usually, about 20 km before finish the team would decide who's the leader (i.o.w. has the best form and most energy for the mass sprint left). the other two (or at least one) will then keep that dude out of the wind and guide him to a good spot for the sprint. There is of course a remote chance that 2 were guiding the 3rd to the sprint and their kinetical energy kept them going strong to get 4th and 5th but that is unlikely. Any ideas here??

There are probably other things, that I may have missed. Please do post your opinions here!

PP: I posted here except on the Cyclingpro forum as this place is perhaps a bit better suited for now.

We're working on logs atm, 1st drafts should be available today. If they're just abit decent they'll be posted on the site se we can all see what happens in a race. Shoudl provide more input about the engine.

I'm not surprised that I find several of my riders in the first group, I've seen it before in a similar fashion. I tried a late escape before the last mountain. Because that's 50 km from the finish that's probably too far, that's why they failed.
What you see here is that the 3 leaders in the first group are domestiques. That's not that surprising, I guess the others are tired. I also noticed in past races that if you want a late escape you can just as well appoint some total crap rider as a leader, because it's a captain who's going to win. Super secret tech.

Why those 3 riders finished in that order, I have no clue. The second in the group is the worst sprinter of the 3, and all 3 of them are better sprinters than your rider.
As to why those riders finished close together I do have a clue: my guess is that there's a piece of code that makes a team stick together, to simulate that they're helping each other. It's pretty clear that they're not, because like you say, if several riders of the same team are in the first group, something has to happen. There's no way they'll finish 3rd, 4th and 5th.

I've seen it worse though, 6 riders finishing together, with 5 of my team. 6 riders in a sprint, and the one guy from the other team wins. Unimaginable.

But these strange results are leading to wrong point distributions. If one of my riders had won, and the others would've been 10th and 11th I'd have gotten less points. It'd have made more sense to finish like that, and IRL it's winning that counts. Obviously the Pro Tour has a comparable point distribution, but it just doesn't feel right. Perhaps there should be bonuses for finishing on the podium, and for winning the rider ranking at the end of the season.

Stephko has come up with a new point distribution that increases the points for higher places. Also he's writing up how match orders work in more detail. I'll check how far he is doing this.

Edit: Just testing to see if I can edit pp's posts.... yup [melihere]

we'll blame it on Meli... shhh

At first glance the result from race 2 makes some sense. A group that finished 21 seconds before the peleton. I'm very curious what's deciding the placement though. My rider that placed second has 1 wheel at sprint, so why doesn't the rider that placed 11th and has 3 wheels at sprint (and is the better rider in every other aspect as well) finish second (for instance)?

Another thing of note, my 3 captains are in the first group, but the leader in the second.

Different winner, same team.[drunks1]

But seriously.

as Shabba said. 11 riders manage to evade the peloton to get a 21" lead and sprint for the end. As massive accidents are not yet implemented (a big crash 5 km from finish could lead to this result) it is difficult to see this happening. the group of 11:

-has 5 (!!!) from Rambobank, 3 from RRR and 3 loners. 3 From one team is very rare but 5? No way any Peloton would let that happen...and certainly now in a flatter course.
-If for some reason a group like this does form, chances are high that one of RRR or Rambobank will win (say 75% Rambo, 24.9% RRR. 0.1% rest). which is just what happened.
-The rider winning, RRR's Tom Rosa has 3 sprint and 4 steer and decent experience, other skills like time also decent. Overall no McEwen at all but certainly not a utter moron either.

In the end hard to say if the result was ok.

But then the follow up: so Rosa has 3 sprint, but does he have the time trial skill (and echelon?) to end up in that group in the first place?

In the second race my best sprint guy did end up first of my group and I asked for late escape with default distances.

Although how a guy can be 0 sec behind and not win I do not understand but...

Also the second guy in my group is 35 sec behind. I guess this makes sense to me as my sprinter left the pack but overall I just don't know enough about cycle racing to know what is normal or not

Well, the other two riders in the group are my best 2 riders, They could easily make it in the top group, with Rosa having a good day jumping along with them. And as said, Rosa has decent skils all over.

@ Mistfit: If I remember correctly, seconds are rounded down. If, sprinting 60 Kph, riders finish e.g. 20 cm apart that is thus denoted as zero secs difference.

Working on the logs has helped us get some bugs out of sprint which should make this order perform better. Still some weird stuff in it, saturday will see another improvement. I just received new logs, will check them and if possible will put them online today. Still very, very crude logs though!

havent seen the log of the protour yet but 21" left over when a group of 11 escapes seems possible to me, no crash necesssary. Early escape could lead to that. That fact that a lot of riders from the same team end up in same escape groups is on our list to look into. Logs and major bugs 1st though.