Résultats les plus réalistes

Any questions related to version 2 or version 3 should be ask here. / Toutes les questions sur la version 2 ou la version 3 devraient être demandées ici.
Post Reply
Richybaby
New in Town / Le Ptit Nouveau
Posts: 28
Joined: Wed Sep 28, 2011 8:16 pm

Résultats les plus réalistes

Post by Richybaby »

Bonjour tout le monde. Ca fait longtemps que je ne me suis pas mis à jour avec la version du STHS. J'avais essayé les version 2.0 et 2.1 du moteur de simulation à leur sortie mais je suis retourné au 1.5 assez vite vu le manque de réaliste.

Quel est la version la plus réaliste maintenant? J'hésite à changer ma 1.5
Foo
The SuperStar
Posts: 1566
Joined: Thu Dec 29, 2005 12:56 pm

Re: Résultats les plus réalistes

Post by Foo »

1.5 sans aucun doute.

la 2.1 même avec des ratings fait pour cet engin va te donner des gardiens sur-humain et très peu de scoring sur tes 2-3-4 ème lignes. C'est un engin pour les vedettes seulement alors que le 1.5 est mieux réparti.
Image
Image
Image
36Henry
The Accomplished One / L'Accompli
Posts: 118
Joined: Wed Dec 15, 2010 1:58 am

Re: Résultats les plus réalistes

Post by 36Henry »

I must respectfully say that Foo has no idea what he/she is talking about. With proper ratings the 2.1 engine is head and shoulders ahead of 1.5 in every possible aspect.

Just a couple of hours ago someone created a topic about how teams wont stop brawling in the 1.5.... :roll:

The myth that only the top line can score in the 2.1 has been debunked several times yet it still lives on because people try the 2.1 engine with ratings that were clearly NOT made for it (or by people who had no idea what they were doing). Tell Carey Price to play goal in the NHL with the goalie pads of a soccer goalie and he wont look very good will he? Or try to run a diesel engine on petrol.

Below is an old scoring chart of the 2014/15 Columbus Blue Jackets. The stats are the averages from five test seasons I ran using the 2.1 engine with my ratings. Ratings that were created specifically for the 2.1 engine.

Image

Does that look like a team where only the #1 line can score?
Image
Richybaby
New in Town / Le Ptit Nouveau
Posts: 28
Joined: Wed Sep 28, 2011 8:16 pm

Re: Résultats les plus réalistes

Post by Richybaby »

Hi 36Henry, can you send the link of your league so i can take a look at these ratings? Thank you!!
Foo
The SuperStar
Posts: 1566
Joined: Thu Dec 29, 2005 12:56 pm

Re: Résultats les plus réalistes

Post by Foo »

Well for a guy who has made 100 seasons testings in pair with other STHS owners I do believe I know what I'm talking about.

It's easy to throw some partial stats here and their and said that the ratings are perfect to sell them 100$ a year.

Goalies ARE superhuman in 2.1 and their is a problem overall with scoring with this engine.

When I do testings, I'm doing it from a player perspective not to have the same results in their respective NHL teams. From what I've seen an analysed, engine 1.5 is the best out there and 90% of the STHS owners will tell you the same. Don't burn a 100$ a year on that with several other free options that will do exactly what you are looking for.
Image
Image
Image
36Henry
The Accomplished One / L'Accompli
Posts: 118
Joined: Wed Dec 15, 2010 1:58 am

Re: Résultats les plus réalistes

Post by 36Henry »

Foo wrote:Well for a guy who has made 100 seasons testings in pair with other STHS owners I do believe I know what I'm talking about.
Evidently not :lol:

You could run millions of test seasons if you want, but if you're basing each and every one of them on bad ratings and you have no idea what you're doing, you get nothing but bad data. That's just basic common sense. And if you don't learn anything from the tests, what are you actually trying to do with them? Is it a case of the blind leading the blind?

Also, 100 test seasons isn't nearly enough to create a worthwhile set of ratings, as Nicklas Backstrom shooting the puck more than Ovechkin in your ratings so beautifully proves. Are you even interested in realistic simulations?
Foo wrote:When I do testings, I'm doing it from a player perspective not to have the same results in their respective NHL teams.
Evidently not :lol:
Foo wrote:It's easy to throw some partial stats here and their and said that the ratings are perfect to sell them 100$ a year.

Goalies ARE superhuman in 2.1 and their is a problem overall with scoring with this engine.
It's also easy, it seems, to make bold statements about things you obviously don't understand.Since you can't do ratings for the 2.1, you conclude nobody can do it. You should read up on Wittgenstein perhaps?

It's also easy to say that 1.5 is the more realistic engine and then proceed to have Nicklas Backstrom make Ovechkin look like a playmaker.

I'm not sure how stats proving your statements to be incorrect isn't enough to at least make you entertain the idea that just maybe somebody knows more about the 2.1 engine than you do, despite all of your 100 (wow!!) test seasons (all of which were spent using incorrect and useless ratings).

1. Goalies are NOT superhuman in 2.1 if you use ratings designed for the 2.1.
2. There is NO problem with scoring if you use ratings designed for the 2.1.
3. Using ratings NOT designed for 2.1 to test 2.1 is like trying to set off a nuclear missile with a simple lighter (fans of James May will know what I mean).
Foo wrote:Don't burn a 100$ a year on that with several other free options that will do exactly what you are looking for.
Show me the free options for 2.1 where the ratings do exactly what my ratings do (or better yet, show me ANY set for 2.1 that does what my ratings do). You yourself just claimed that ALL goalies in 2.1 are superhuman, and that there is an overall scoring problem with the engine. Neither of these things exist in my ratings, so where are these free options where the goalies are NOT superhuman. And if such free options that work with the 2.1 engine exists, where as in mine the goalies are NOT superhuman, why are you still claiming that goalies ARE superhuman?! :?

As for having my ratings for sale, yes I have. I'm not forcing anybody to buy them though. I also see you have a banner linking to a site where you sell your ratings for $5, last updated in 2011. Having looked at the stats in your league I can see why your ratings are free these days.

Ultimately, people get what you pay for. It's very much a buyers market and if they don't like your product, or feel it's not worth the price, they'll simply not buy it. However, it might interest you to know that everyone who has bought my ratings have bought them again for the following season. Would they do that if the ratings didn't deliver everything I claim they will? Doubtful isn't it?
Image
ynohtna
The Addict / Le Drogué
Posts: 770
Joined: Mon Aug 14, 2006 5:15 pm
Location: Vancouver, BC
Contact:

Re: Résultats les plus réalistes

Post by ynohtna »

I have to say, our foray into 1.5 so far (pre-season and first week into the season) has not produced the most appealing results yet though it is a small sample size.

My worst fear going from 2.1 to 1.5 was that unmanaged teams would win more than they should and it's happening.

Maybe I'm biased and really haven't quite given 1.5 a full chance yet.... but aside from crazy scoring, 2.1 sim engine is really a superior SIMULATION.

If you only look at end result and box scores I guess 1.5 is what you can call realistic... Super goalies aren't great but super scoring lines, I've always attributed it to those GMs doing a better job of managing their team and figuring it out. Smaller league, more talent on teams...

I'll still keep an open mind for a bit more but if you want to reward GMs that are actually trying and using strategies... 1.5 is not the right choice.

We'll see, I see a few of my GMs are changing lines a lot. Hopefully it turns out ok and I don't have to do anything drastic.
BFHL Admin/Commish
http://www.thebfhl.ca/bfhl
Now in Season 15, using BRHL Player v2.1 Ratings modified.
Foo
The SuperStar
Posts: 1566
Joined: Thu Dec 29, 2005 12:56 pm

Re: Résultats les plus réalistes

Post by Foo »

36Henry I just saw you private message, first of all thank you it looks like very detailed.

I'll get back to you on it when I'll have 2 minutes to analyse it.
Image
Image
Image
Foo
The SuperStar
Posts: 1566
Joined: Thu Dec 29, 2005 12:56 pm

Re: Résultats les plus réalistes

Post by Foo »

Wow, great email thank you for taking the time.

Still, it is not a fair fight.. If we take Backstrom for example, in your testings he plays with 2 great wingers.. in my league he is all alone and play more minutes everywhere.. would be kind of stupid if he scored just 20 goals... Players will produce differently on how they are used in the simulation.

Secondly, You have access to all my stats, I don't.. easier for you to point what's wrong with 1.5 then reverse.

I propose you send me your last year ratings I'll do a full season simulation with your settings and output the results to really compare orange with orange here. I promess, if I'm wrong I will retract but I will not pay 100$ for your ratings, still plenty satisfied with mine. ;)
Image
Image
Image
36Henry
The Accomplished One / L'Accompli
Posts: 118
Joined: Wed Dec 15, 2010 1:58 am

Re: Résultats les plus réalistes

Post by 36Henry »

Foo wrote:I propose you send me your last year ratings I'll do a full season simulation with your settings and output the results to really compare orange with orange here. I promess, if I'm wrong I will retract but I will not pay 100$ for your ratings, still plenty satisfied with mine. ;)
Haha, check your messages. Not quite what you asked for but enough for you to make a fair evaluation :lol:
Image
Foo
The SuperStar
Posts: 1566
Joined: Thu Dec 29, 2005 12:56 pm

Re: Résultats les plus réalistes

Post by Foo »

fair enough.. I'll analyze all this and get back to you on this one. Thanks!
Image
Image
Image
Post Reply