BuzzerBeater Forums

I feel your pain.
I can log in everywhere using my access key, but when I try to get into the bb-training site this is exactly what happens.

I have been able to get into the site once to update my data, but never since then...
It's starting to bug me seeing as the data was looking so awesome, and I bet its even better now.

you can try to use something like this : http://anonymizer.nntime.com/ or http://daidostup.ru/ In my case this work perfectly. But, you will can't sent training data:-/

That happened to me until I got the update which is found on the main screen on point number 2. Click the word "here". After I updated it worked fine.

hi,

last 2 weeks i always get the message that sending is failed, but at the page everything is updated!? not sure if it's really is, but is seems to work well for me...

That happened to me until I got the update which is found on the main screen on point number 2. Click the word "here". After I updated it worked fine.

+1

Thanx! i had the same problem too, but this solved the problem :)

Question:
If we go to, Roster, pick a player, and then we click on Share this player.
These are the players that help the project, or those player are all that have training during the week?

Last edited by Seba at 2/17/2012 7:33:16 PM

Question:
If we go to, Roster, pick a player, and then we click on Share this player.
These are the players that help the project, or those player are all that have training during the week?

Share the player just creates a link to that player that anyone can see. By default only the owner of a player can see them. Just a way to facilitate discussion of players, and you can use it with any player. Every player on your roster is being included in the data side of the project, so you don't have to explicitly share them to include them. Does that answer your question?

Does that answer your question?

Very clear!
Thanxs!

My apologies for being delinquent in adding new things to this. I've been planning to release the trainer level one for a while, but I've been waiting in hopes that the data would improve. Alas, even with close to 500 users across the US and non-US sites, well over 5000 weeks of data and 10,000 pops, the numbers are still messy. So, I'm releasing it anyway, and trusting you all to take it with many grains of salt!

If you log in to the site (http://bb-training.comuf.com/club/) and go to the analysis page, you'll see "Trainer level analysis" as an option.

I've allowed you to select the age range of players to include. There's a trade-off here, of course-- the smaller the range you set, the more consistent the data will be, but the less of it will be included. But, it's a useful experiment to play with it a bit to see how consistent the results are depending on the input data.

Part of the reason for doing this is that the results, especially for level 6 & 7 trainers, is really bizarre. Level 6 is almost always slightly *slower* than level 5. And level 7 almost always looks like a bigger effect than common sense suggests. I strongly suggest expanding the "raw data" tab and looking at the actual rates being used to generate the plot. Especially for level 7, you'll see that there's a very small number of data points, so you may want to mentally exclude those. It's also useful to see that even with what seem like large samples, the data can still be funny (for PG Pressure training, which is by far the most common, level 6 has produced pops at an almost 5% lower rate, despite having >300 weeks of data for each for just 18-20yo's).

So, take a look (as long as you've contributed at least 4 weeks of data) and draw your own conclusions!

Level 6 is almost always slightly *slower* than level 5.

Bwhahahahahaha

- L5 or less trainer owner

Level 6 is almost always slightly *slower* than level 5.

Bwhahahahahaha

- L5 or less trainer owner

Ha.

Lest I be misinterpreted, let me make clear that I don't *believe* this result. In fact, the error bars show a pretty big range of possible values, including values where (as we assume the true value is, given our belief in competent, non-malevolent, BB's!) it's slightly faster than level 5. What I'm saying is that *we can't tell* what the real difference is.

But I think it's safe to assume that if we can't tell how much of an advantage it is (or whether it's an advantage at all) with this much data, the effect is likely to be pretty small.