BuzzerBeater Forums

simpler system:
3 or more player - old code
less - wo
new code is unnecessary

Last edited by darkonza at 10/16/2014 3:12:43 AM

@Yuck: OK, no problem, but what exactly do you mean by lineup blueprints? Give me an example, please. It may be a stupid question, but that might be due to exhaustion from the offseason...

@darkonza: even more walkovers? I don't think that's the solution. It's easy, but not a solution.

I apologize if I was vague. In my mind I was picturing something like this...

Test Sample #1

PG - Player A / Player C / Player D
SG - Player B / Player C / Player D
SF - Player E / Player E / Player E
PF - Blank
C - Blank

Then have a list of teams sign up that are willing to set their lineup for the balance of their PL games this way. Some using LCD, some Coach Picks from Depth Chart and some SFDC.

Test Sample #2

Similar layout to above but obviously different enough that you could draw information from the results. Same as above, a set number of teams willing to operate under these guideline within their PL schedule. You could have as many test samples as you would find relevant towards achieving your goal.

Double up the PL games for this season? May not be possible to code but say for instance you and I had a PL match scheduled for tomorrow. Double that on the schedule. You and I would set two lineups for games tomorrow. Testers would be able to use the same lineup and settings but could change the coach's decision. You would have a viable test sample comparing two games with identical game shape (which affects substitutions), same enthusiasm levels and common opponent with those same variables.

No matter how you look at it, I feel your pain. This is a massive undertaking but there are so many variables that I dont know how you can collect accurate data unless you are running the same variables as you tweak the code. If I run a lineup for you tomorrow, I could have completely different results in two weeks running the exact same lineup and settings. My game shapes will have adjusted up or down, enthusiasm has increased,and training for some players make them more valuable to the coach than they were previously. I may have transferred some players between now and then, I am probably not playing the same opponent, but if I was, he will have the same variables hit in a few weeks too. We just can't draw accurate conclusions from running a series of tests week to week using the PL format.

I have looked at it pretty closely for some time. I came to the conclusion a while back that it just wouldn't be possible so I did not participate in the testing. I also, childishly, despise the thought of getting rid of blank and instead of being constructive and helpful, I have been more of a hindrance. For that I apologize.

I fully agree with this, I would force users to input 5 starters, no sending orders without full starting lineup.

Backups can then be auto-filled.

Why not just preventing the users to set a lineup where all the selected players have not a position...?

@JoeMaverick and other: Because we already have lot's of default lineups set that way. We'd have to fill them automatically which is not something users would like (forcing something on them). Also, I believe more resourceful users would find a way around it somehow... But, I will give it some more thought, maybe there's a way to make sure.

@Yuck: No worries, I love to hear opinions, even if they don't align with mine. And no, you haven't been a hindrance, just a part of a discussion. As for the test sample, I see your point and will probably do it at one point, but for now, I'd like everyone to just use a completely blank lineup one week, and then a completely full one the next, LCD. Doesn't have to be in that order. Having two private league matches is a very complicated, so not an option.