ACC Monday nights Season 2 Schedule, Entry List, Standings and Sign Up

Will need all info to enter you as it says in the first post here.

real shame race performance and pace hasnt been considered. Just because somebody can smash out a good time in quali doesnt determine their pace across an entire race. im just a good hot lapper and the qualifying track was 1 i dont know at all.

most races ive had real good fights with other cars for the silver win. Bu3ller at misano was a classic battle that went on most of the race and was only decided by traffic. all season ive been having great fights with the other top silver drivers. So i guess that will continue except ill be in a different class to them.

The main idea for the classes is to give more people a chance if silver is flooded with guys that are well within the Pro qualifying range it doesnt really give those that have a true pace in that range much chance.

However even with 1.5 seconds between classes its still tough for those at the bottom end of each class as obviously they would compete better with those within .5 of their times and these guys may be in the lower class. Its never going to be perfect due to this, with lots of new people joining we can only go by their qualy times then adjust if needed during the season to try avoid people well outside the pace ranges of the lower classes to dominate them.

Your pace however in both qualy and races reflects its generally better then 1.5 of seconds off the top times. The qualy server isnt a track anyone was familiar with, it was decided by a poll. I would have prefered a track that everyone knew but it made for an interesting challenge, those that had more limited time to set a lap and never drove the track would likely not have shown their true pace.

thats because a linear time for class split is impractical due to driver skill levels and laptime consistency differences between somebody whos within 1 second of the fastest guys and somebody whos 3 or so seconds off.

A non linear spread is always possible and everything is up for discussion or adjusting. We only have 3 classes available in ACC with a reasonable spread of pace in the field so need a bit of room in each class.

The main reason in season adjustments were needed is because a bunch of people that qualified 2 to 3 seconds off pace ended up being .5 to 1 second off in actual season rounds which is defined as a different class in the way they were set this season.

There are no tools in the game for this and we are just blokes with limited time trying to do our best - every tool for collecting/analysing the data is something that we have built (and are still building) and some of this is being done manually by already busy admins like Marty who have a lot on their plate just keeping the servers up, set up correctly and doing all of the forum admin stuff. It is hard to get a good basis for deep statistical analysis because the availability of data and tools for breaking it down are not so great.

The way we have the league set up, we have to choose a way to measure it that is easy to set-up and track, and gives us a way to measure new drivers quickly (in one session). We also need a simple benchmark for ongoing adjustments like those discussed above. It is not ideal/perfect, but until we have a way of pulling lots of data and parsing through it easily, this is the compromise.

In addition to that, the quality of data available to us is also not good/consistent - there are variations in the car BOP and physics engine, session set up and weather and so on that all add variables to race data that are difficult to control for. There is no record of track conditions within the .json files we get from the server, there is no versioning information re: the car BOP or physics engine that we can use to filter the data unless we create a method to track it ourselves.

In an ideal world we would select a pool of race data from the most recent race on each track, using only one session where all drivers were using the same game version, BOP and session conditions, and then take an average of say the best 5 to 10 laps for each driver. We could then create a 'relative pace' value measured against the fastest driver for each track and combine them into one to act as each drivers own total 'relative pace'. That is a lot of data to crunch for 40+ drivers - we are on the way to having tools to do this but we aren't there yet.

There are much better ways to set classes and judge driver strength, but we need something practical we can work with today - also congrats on pro u fast git.

Also remember it's GT3 so if anything doesn't go your way just blame the BoP

I'm the opposite - I qualified Silver but am now struggling in the AM class!

Is it your general pace or more bad luck with contacts etc? Unfortunatly thats something that is very hard to account for and is more likely the further back anyone starts of being caught up in early incidents. This will generally also slow down race pace due to early damage and maybe stops out of sequence and maybe being stuck in traffic racing for position with cars that may be a bit slower in single lap pace.

After this season we will take a look at class ratings and see what can be done, it will never be perfect but could be be improved possibly.

This season with 1.5 second gaps between the 3 classes its still a fair gap between those at the top of the class and those further back. With a few Silvers that were running more mid field Pro pace it made it even harder for the Silvers closer to the top AM pace.

With only 3 classes available in ACC we can set and a general pace gap of about 5 seconds from the front to back of the field. Trying to find the best way to split everyone is always going to be a bit tricky.

There is also the extra variable of base physics updates, some may fond more pace as the cars update and others may struggle more to come to terms with the changes. This makes it even harder to manage and try to keep the classes more competitive as the season progresses over time.

TBH, I don't really care about classes, so long as I have some other cars nearby to race with, and to try to better my own personal performance. But that might be just me.

living abroad; its very nice to have you bunch of aussie blokes to race against and talk a little sh!t.. yeah there are a few poms and kiwis and it's great they are aussies too as we are all mongrels at the end of the day, especially me...

In the end the more people that race the better for all, the classes are just to help keep some motivation to the guys a bit further back. It really shouldn't be enough to make people choose to race or not, after all the prize money is the same in all classes and in all positions.

Getting bigger more competitive fields will make for more chances of better racing in the field, it also makes it more likely to be involved in some early carnage but thats all part of racing. Just doing my best is all I try to do too, the harder the challenge the more rewarding a decent result is. Even being stuck behind the @tezpez85 train was fun in its own way and Im not the first passenger on that line this season.

Good idea.

Its an interesting idea, one thing with race pace however is there are many factors here that can make people slower then they would be. If your in a battle or get early damage and are forced to run a non optimal strategy you can easily be 1 or more seconds per lap slower then you would with your planned strategy and clean air.

It would be good to find better ways to grade people, also BOP per track in certain cars will tilt performance one way or the other. I think qualy pace and best race laps are probably a better indicator but then best race laps also will depend on how the race is going. If you need an early stop and need an extra stop later in the race you will likely have a better car then everyone else when the track is at its fastest. You could set lap times considerably faster then in a normal race.

I really don't know the best way to set classes, more discussion on this in general is always welcome. We can always try different methods, but it seems there will need to be certain criteria set during a season that will adjust peoples class if they clearly break out of the one they are in.

A qualy server is a good start for a base, but its not going to be a perfect indicator ever.

Maybe there are two ratings we can use and blend them - hotlap pace and race pace

Yeah it's obviously difficult controlling for so many different variables - some drivers may always be able to maximise (or close to maximise) qualification, and then fade in terms of race pace, others vice versa; and others' pace may be hidden because whether through bad luck or other factors they are getting involved in incidents by themselves or with other drivers. Similarly, progress in races is also likely to be affected by other's people incidents and so on.

And then there will always be people who are peaky, or inconsistent or find themselves on the borderline between classes and therefore difficult to sort.

As has already been thought of above, ways to take into account more variables is probably the best way to go - though from what @SteveDrivingSlowly has already said, that sounds potentially very complicated!

I do not envy coming up with a fair and robust way of assigning and then adjusting class membership, when we have grids of 33 and about 40+ active drivers so, er, good luck I guess

it shouldn’t be toooooo hard to do but there are a number of other chunky things I need to do with the results site before I get onto it - we’ll get there!!!

"firstName":" Brett "
"lastName":"Maybery"
"shortName":"BMR"
# 155

also looking for a team mate to do the endurance races with.
I am no professional typically drive Porsche but happy to work wit somebody.
drop me a PM