i agree with WMJ. whoever wrote the rules didn't really understand the game that well and also there was missing knowledge (these rules were made before AGG was discovered)

going +3 is only artificially handicapping some tracks with no gain at all. this isn't like non-SA or non-custom play where the strats change a lot due to a restriction. here it's just plain slower to use those settings with no gain.

i vote for starting at 0. it would account for all the actual jumper levels.

 It should be noted that +0 is around the threshold where you can accelerate above Blood Hawk's unboosted max speed while drifting (haven't really tested this that much though). Further testing could be done to ensure that wherever we decide to cap "jumper" settings is faster with optimal jumper strategies than terrible drifting at +0 settings. I myself am too lazy to conduct such tests and believe it is worth it to allow people to use anything right of the graph to allow for much better floating to be achievable.

I also think we should archive all current jumper ladder times into a "pre-2018 rules" forum post to ensure that times that required hard work on awful settings (mostly thinking of Linner's Sector B 1'04) don't get forgotten.

Yo.

Alright, so here's my attempt to clarify where sliding/drifting stops being beneficial over regular driving. First test is neutral settings (+-0) on PT1:

youtu.be/HDgqxr7SHUQ

Screencap of the last frame (Left = drifting. Right = normal driving. This will be the case for future tests and screencaps) :

In the above PT1 demo, it was slower to slide/drift around the first 2 hairpin corners after the chicane because the benefits of drifting are weaker the faster the ship is travelling around the corner and the actual drift itself is usually slower than taking the turn normally - what makes it faster overall is the higher exit speed, which can only be properly utilised on a straight that's at least somewhat lengthy. Even though these downsides greatly limite the effectiveness of drifting at neutral settings, I was still able to save almost a whole second on the opening lap on PT1 in the TAS demo.

The benefit is also greatly increased when longer corners and boost are thrown into the mix:

youtu.be/PQXSbMJXLCM

The Silence 2 scenario is probably the best and easiest way of testing where the speed benefits of drifting stop so here's the same test but at +1 settings:

youtu.be/oOOzd-apUg0

Despite the settings now being above neutral, drifting is still faster by 0"62...somehow.

+2 settings:

youtu.be/aqJxdTtsuTk

Driving normally at the start saves 0"05 and only doing one drift on the first dash plate saves about ~0"2 over doing two of them at each dash plate.

This is where it starts to get tricky. In my opinion, I think that the 0"04 / 2-frame difference in the +2 settings test is negligible since the execution of the drift needs to be near-perfect in order to gain a tiny advantage, which, at least in this scenario, I think is only realistically achievable in a TAS.

Ultimately, these are (somewhat rough) TASes and not real-time runs so the potential to cut down time by sliding/drifting may not fully translate from these TAS demos to console runs, but having said that, the fact that there's potential in the first place is concerning. Despite the benefits of lowering the minimum settings allowed to 0 or maybe +1, I'm currently leaning towards slightly lowering it to +2 purely because that's where the sliding/drifting benefits seem to stop, at least for console runs.

how about a machine dependent cutoff

lets say you can use +0 but not with blood hawk or any E grip machine

 This is a very interesting comparison. I think it makes sense to see these differences in the drifting cutoff because of the way weight of the ship affects the grip (and drifting ability). 

A thing not to overlook is that if you use +0 as the setting for grip driving that is in the vast majority of cases a very bad setting to go with. Most tracks the optimal grip driving settings are actually around +5. If you pick the optimal grip settings on Port Town and Silence 2 I'm sure the grip driving would totally defeat the +0 drifting strategies. So, in practice you would never use a +0 drifting strategy. The +0 setting is mainly useful for floating strategies such as the loop skip on Sector Beta. At +3 you decellerate much faster during floating and thus the loop skip is much harder. 

I am sitll in favor of the +0 cutoff for above mentioned reasons.
 

That's a fair point about +5 (or thereabouts) grip driving being better than +0 drifting. I'll concede to that (especially after doing a rough test or two just to be extra sure) but there's one issue that I'm still not sure about: how do we deal with drifting on tracks like Sector Beta where +0 is ideal for separate reasons such as better floating? Even if we add a restriction like dkr_paddy suggested where E-Grip ships are forbidden from using +0, the Twin Noritta (C-Grip) is still able to save an extra 0"1 or so by drifting on the first corner of Sector Beta using +0:

youtu.be/Itn-VtBBT4I

 can't we do like the GX ladders and restrict the ladders to only 1 style? like, they can't snake in 100% even if it gives some ships speed and they can't submit max speed times to the snaking ladder even if they're faster than their snaking PB.

i guess the 0+ thingy with TN would still be tricky but that track could be the exception, don't you think?

 anything else we'd need to know and test before changing the rules? so far most votes are in favor of 0+

anyone we still need to ask before going forward?