Subject: Questions & Answers

You seem inconsistent on this issue though. Sometimes you say you agree with unexpected results, saying that's hockey, things do happen. Other times you say that it is ludicrous, insane and should not have happened that way at all.

To use your example, the OHL team has very little chance of winning. But they do have a chance. Think of this as a world in which there are 1000s of NHL caliber teams playing 1000s of OHL teams each day (again just using your example) are you saying 1 out of those 1000 each day, the OHL team won't win?

So it may seem like "a lot" but, we also play a LOT more games in "our" world. So if every day 1 of those 1000 games ends up in an upset, that is too many? That is still "hardly ever" but all we see in this thread are the 1 of 1000, never the other 999 games

yes.. you were right ! , and it is very hard to accept when this 1 hits you. I speak from my experience.

i still hold that at the moment the difference between teams is quite small compared to what they would be when the top teams are at ~250-300 team strength.

I dont see a 30 strength difference as that big. 100 now that would be a different story.

Not really inconsistent.

I have said that yes, weaker teams can beat stronger teams, but i also said that PPM needs to reduce how often this happens AND how drastic those results are.

To use the OHL/NHL example, an OHL might beat an NHL team on a lucky day, but they probably wont win 5-0.

but how do you know how often it happens? How many games are played every day on PPM. Out of those, how many "weak" teams beat "strong" teams, and by drastic results?

No one has those #'s, all we see are the "unexpected" results, while 99.9% of expected go unreported, because it's "expected"

How about - from 10 games in one day from any league, most common result are 2 games with unexpected results. Of course not like ZenekKonevka example, but still unexplainable.

20%?... I don't think so

You don's have to, I think so.

)) ok, I think you're wrong.... my guess is 5%.... now, who has enough time to verify all the yesterday games, and just leagues I and II... the III, IV ones have too many TNNs (except Slovakia and Czech Rep)

One guy from Serbia is doing analysis from his league (don't know why he needs to bother with that s**t at all ) from day 1 until now. He has plenty of credibility to believe him, and I do. There are small random, random, and big random games as he called them, so 2 per day is most often as I saw.

I wonder how would this guy from Serbia classify those 2 yesterday's mathes in our league:

hockey.powerplaymanager.c...
and
hockey.powerplaymanager.c...

This kind of upsets are very often in most of our league days during this season. Previous seasons I didn't see that many unexplainable game results, IMO.

Just to refresh memories.
hockey.powerplaymanager.c...
This actually happened not once in 1000 or less. It happened 29 times in 500 simulations of the same match. So I know pretty well, how often it happens and so do the others. It's way too often. Period.

You can't say 30 strength difference is not big. It's huge, if you compare strength 20 to 50. And it's smaller if you compare 120 to 90, but it's still 33.3%. To put it another way, relative difference is what should matter, not absolute.

Absolutely. How do we know how many? We don't. We do know that the number could be too high though, even Vlady has explained that already, and PPM are going to be looking to make some improvements in the off-season.

its only those kind of percentages if you think that it starts from 0 base.

Would you think a team with 2 overall strength is twice as good as one with a 1 overall strength? Or would you say its too close to differentiate them?