BuzzerBeater Forums

Tough question...

I understand why they want to make changes - it's a bit off balance at the moment and changes are needed to make progression for the game, so I support that.

The concept of how to do that - decrease something, increase something, some combination - I'm not sure what would be the right approach, I am sure there are insane amount of data and simulation capabilities which can give the right answer to the question.

BUT, what I absolutely do not agree with to release something in a rushed pace like this - yeah we will do something in s50, will release it to some games for testing and live from s51 - that's disrespectful to the managers who built teams consciously... at least 1-2 seasons of notice would be expected.

The app question - in my view that's a convenience thing - but a very popular and demanded one from the community for sure.

I think I would do the following:
1. Complete dev for the engine, implement into private leagues and scrimmages for testing - but do not implement in s51, only in s52 or s53.
2. Do the app and release it as soon as you have it ready.

I totally agree with you.

The case you present is exactly mine, and i understand that of hundreds of managers... 9 seasons training my players thoroughly and now that i have to get performance, they change the formula to make external tactics more efficient... these changes should be announced within 3-4 seasons.

I think the change is great and i understand that it is necessary, but obviously it should be announced with much more time so that the teams already formed can change their players with more patience.

What I mean is that with less impact (which is not necessarily a bad thing as you don't want to overcorrect), the need for far in advance warning diminishes.

The first thing they should do, which is also VERY easy, would be to address open shots and value them against a constant: say the equivalent of a team with skills equal to 7 all around, or 8 or 6, you get the idea. Then higher salary players will shoot better than scrubs on open shots (realistic) and flow will finally matter more as it does in real life. I have no problem with players hitting 50% of their open 3s and 25% otherwise. It will be quality over quantity quite literally and it's possible that princeton and low post may become more appealing compared to the fast pace counterparts.
This change should be easy enough to make as it only involves modifying parameters in a calculation step and nothing in the actual GE that simulates what happened up to the point that the shot was taken.

Of course they could also look at how OD, ID and SB are used by the GE to affect shot selection, distance and the outcome of the shot (score, miss, block and foul) as some rebalancing there may be warranted, although I think this would take a lot of time and a lot of fine tuning and some stuff may not be easy to implement. In general terms, you'd probably want to feature SB more in various calculations, specifically for jump shots and outside shots and possibly value it against normalised values for guards (since SB is not a guard skill and is difficult to train for short players, you can't expect to have 20SB on a guard for it to be meaningful). At the same time OD's impact could be reduced to compensate.

Finally I would like to see more of the help defense event (where a player rotates to alter a shot), but not sure it is feasible. This is event e0003t0, just to be clear, and since all assist events start with e0001 and all the defense events with e0002 it stands to reason this is something that was added later and/or which is calculated separately from normal individual defense, but it's probably part of the core of the GE, not just a simple calculation.

Last edited by Lemonshine at 6/10/2020 9:07:12 AM

Also we know JR won't solve ANY problem on its own. Aleksandar had those 2 players he bought for 11m and 13m respectively who had (I'm going by memory here) 20/20 and 18/18 JS/JR:
Siber (33041386) and Klavc (34962966)

The first guy (who I believe was the 20/20 guy) shot 25% from 3 in B3 games (best seasons 34%). The second guy shot 29%. These 2 were NT players with high salaries and played together while he had them.

You can lower the cost and cap impact of JS and JR all you want, but it's not going to matter if you need 23/23 to shoot decently against high OD. Increasing OD cost and cap toll would be a bit better, but it's unlikely to fix shooting percentages (unless you make OD's cost so prohibitive that players will all have 12 OD and 20 JS and JR).

Probably we should look at some competitive D4 and see what it looks like when skills are lower and closer in value: if anyone can suggest a D4 league with no tankers and where the win number between top and bottom was close, please let me know. I can run numbers on the boxscores for an entire season.

Last edited by Lemonshine at 6/10/2020 11:54:09 AM

I used princeton a lot with average skilled players (no elite skills, but also no holes) in the first few seasons of utopia and it was fine for me, so you may be right and that's why I think we need to check both sides of the spectrum (and I might do this when I get supporter in a week or 2 and I work on collecting pbps), which is why I mentioned a competitive D4 with no tankers and no bots.

Anyway, if you fix open shots the way I said, lower leagues would be largely unaffected while top players should indeed shoot better. Think about it: a D4 player is likely to face a team made up of outside players with 7-8 OD already, but that number for D1 is probably more like 17 or higher at the guards slots. So when you pit open shots against a constant rather than some illogical value created from the opponents' OD, it stands to reason that the biggest impact will be where skills are higher. In D3 and D4 there is more balance because: A) salaries don't vary much between different tactics and B) there are fewer extremes (eg. guards with the equivalent of 18 IS or players with 30 TSP more than the league average for a starter).

Using team OD to calculate the expected FG% of unguarded shots is complete and utter nonsense and if this is indeed the way it works, then it should be changed first. At the moment it's like you are double counting OD in the defense: if the shot is defended, then OD is used, if the shot is not defended (so OD failed to stop a good pass and to close in on the shooter), then OD is used to calculate some value to oppose to the offensive skills relevant for the shot. Why??? The only problem I see with this is that an offset patient would become even more broken as more unguarded shots would go in than right now (at the top).

Another thing I never did was to check if different shots have a different block %. This is another thing that is powering LI: while in the real world layups, finger rolls etc are efficient shots but also more prone to being blocked than, say, 3 pointers or jump shots, especially for shorter players. My suspicion is that in BB there is no significant difference in the probability of being blocked other than whether the defender is a big man or a guard and that layups from the guard slots may be in fact more efficient than layups the big men slots, which is counterintuitive. One thing at a time though.

I will run some numbers on boxscores later this week so we can appreciate the difference in other aspects (namely pace and most importantly possessions) between tactics.

Last edited by Lemonshine at 6/10/2020 12:09:25 PM