Is Preseason Prediction Ever Accurate?

Chicagofan76 · 01-17-2012, 06:03 AM

Last season my EBL Preseason Prediction had my team winning 128 games. I won 105 of 144. This year they have me at 98 wins. see the rest below. Strategy and finances are set to 1976.

1. Who pays attention to this feature?
2. Have you ever seen 1 spot on?
3. Post yours.

hfield007 · 01-17-2012, 06:14 AM

I don't want it to accurately predict the championship which I'm glad it doesn't.
I do check it each season.
It does give a very accurate prediction I have found for me in guessing who the top 3 teams will be usually in each league. (with a few misses here and there which I like).

BusterKing · 01-17-2012, 08:27 AM

Wow, with that kind of record it seems like your team is stacked!

You actually enjoy the fact that you are so dominant over the other teams and lack of competition?

Just curious, because I personally like to make the game harder for myself and impose a lot of house rules so that I don't take advantage of the weak AI.

mouse001 · 01-17-2012, 08:31 AM

I think there's enough fog of war in the preseason predictions. If you have a big enough league it's good at seeing whose the top 50%, whose the bottom 50%. With plenty of room for surprises of course

Markus Heinsohn · 01-17-2012, 10:25 AM

Quote:

Originally Posted by BusterKing

weak AI.

What weak AI?? I always suck even without any house rules. Maybe I'm just a bad GM

sc_superstar · 01-17-2012, 10:35 AM

LOL @ Markus

Bluenoser · 01-17-2012, 10:37 AM

1. Who pays attention to this feature?

I do.

2. Have you ever seen 1 spot on?

No, and I hope I never do, because then I'd believe there was a problem with them.

Jaxxvain · 01-17-2012, 11:43 AM

I haven't seen it "spot on", which is good. I think it has a good amount of "fog of war".

But I do wish it would do a better job with the "realism" of the prediction. If you notice it often lists the SPs as having 40+ starts and the hitters almost always playing 160+ games (and seemingly not ever getting subbed out).

joefromchicago · 01-17-2012, 11:53 AM

In the season I just finished in my dynasty, the AI accurately predicted the pennant winners in both leagues, and even got the runners-up correct. That's unusual, but not unprecedented. What I've found is that the AI will sometimes fixate on a team, predicting it to win the pennant for three or four straight years, even though the team never makes it that high. It's like the AI is saying "OK, this is the year, I just know it!"

Biggio509 · 01-17-2012, 12:33 PM

I think this is the standard formulas that stats guys use to predict teams' records in the real world. It is possibly something like this.

Pythagorean expectation - Wikipedia, the free encyclopedia

A Pythagorean expectation could be set as a function of all players' past runs and runs allowed. The Pythagorean expectation rarely nails the wins correctly but it is used as a measuring stick to determine if a team is under or over performing. For instance, I have read some article claiming the Houston Astros were over achievers in the 1990's to 2005 because the expectation was lower than actual wins.

Erik · 01-17-2012, 01:30 PM

Is there a way to pull this report up mid-season? I ask because I just started a new season and forgot to look at this report during Spring Training. Now when I go to Reports and try and pull it up it says it doesn't exist and I have to wait for next Spring.

I thought I read somewhere this report would be available during the entire year so we could reference it? How would I go about pulling it up?

Thanks.

Bluenoser · 01-17-2012, 02:34 PM

Quote:

Originally Posted by Erik

Is there a way to pull this report up mid-season? I ask because I just started a new season and forgot to look at this report during Spring Training. Now when I go to Reports and try and pull it up it says it doesn't exist and I have to wait for next Spring.

I thought I read somewhere this report would be available during the entire year so we could reference it? How would I go about pulling it up?

Thanks.

If you run the report right after ST, then bookmark the page, it will be there for you to view at anytime during the season. The key is, running it before the season starts.

I always like to wait until opening day because then it covers opening day rosters. Running it prior to or during ST doesn't give a true picture imo because you have more than 25 guys on your active roster and injuries happen during ST that may not be accounted for if you run it before ST ends.

mouse001 · 01-17-2012, 02:47 PM

Quote:

Originally Posted by Bluenoser

If you run the report right after ST, then bookmark the page, it will be there for you to view at anytime during the season. The key is, running it before the season starts.

I always like to wait until opening day because then it covers opening day rosters. Running it prior to or during ST doesn't give a true picture imo because you have more than 25 guys on your active roster and injuries happen during ST that may not be accounted for if you run it before ST ends.

You do have to run it in the preseason for it to appear though. Pro tip: Run it Opening Day, before any games get simulated. Uses any updated rosters and such.

MizzouRah · 01-17-2012, 04:06 PM

I think the AI is quite challenging.. year 9 of my franchise and I have yet to make it to the WS. Won the WS with every verison previous to v12.

Like most, the report gives you a good idea of who will be good and who will be bad.. but like real life MLB preseason magazines, I don't want it 100% accurate either.

Run it after ST and you won't have to bookmark it as it will be under the reports section.

simcrazy · 01-17-2012, 06:31 PM

Quote:

Originally Posted by busterking

wow, with that kind of record it seems like your team is stacked!

You actually enjoy the fact that you are so dominant over the other teams and lack of competition?

Just curious, because i personally like to make the game harder for myself and impose a lot of house rules so that i don't take advantage of the weak ai.

+1

Chicagofan76 · 01-17-2012, 09:12 PM

Quote:

Originally Posted by BusterKing

Wow, with that kind of record it seems like your team is stacked!

You actually enjoy the fact that you are so dominant over the other teams and lack of competition?

Just curious, because I personally like to make the game harder for myself and impose a lot of house rules so that I don't take advantage of the weak AI.

This league is a fictional Historical recreation. My Endor league i have been runing since the mid 80's when i go tired of playing with Star Wars & GI Joe figures and turned them all into baseball players. So I am still in teh dominant part. In 2009 I move to Rock City. I will have 3 decent offensive players but no pitching and no budget to work with. My league really starts in 2020 when I played as the new Ford Thunderbirds GM winning 3 WS in 7 seasons. Glenbard I had originally whether it was thru strat 0 matic or Hardball, TLR3 or HH had originally won 20 consecutive division titles not winning a WS until 2023 and never has made the playoffs since. They won the 2004 WS. (I originally had 4 2 team divisions in 2003 4 3 team divisions 2004-2008)

With this recreation, when I get to 2038 the last season I played on HH2k3 I will dump my 3 boxes worth of stats and print out the history of this league and use that for any future recreations of the EBL.
and my team doesn't seem as stacked as i would like them to be. I suck on D and have a terrible bullpen.

Chicagofan76 · 01-17-2012, 09:15 PM

Quote:

Originally Posted by Jaxxvain

But I do wish it would do a better job with the "realism" of the prediction. If you notice it often lists the SPs as having 40+ starts and the hitters almost always playing 160+ games (and seemingly not ever getting subbed out).

With my 4 SP I skip my #4 5-10 times so my top 3 SP see 38-39 starts. This prediction didnt even take into account the 3 NL SP who were injured as of OD and will win over 18 games.

Chicagofan76 · 01-17-2012, 09:16 PM

Quote:

Originally Posted by Markus Heinsohn

What weak AI?? I always suck even without any house rules. Maybe I'm just a bad GM

they prediction has you going 19-9 you went 19-11 last year a clone of Greg Maddux.

Chicagofan76 · 01-17-2012, 09:17 PM

Quote:

Originally Posted by joefromchicago

In the season I just finished in my dynasty, the AI accurately predicted the pennant winners in both leagues, and even got the runners-up correct. That's unusual, but not unprecedented. What I've found is that the AI will sometimes fixate on a team, predicting it to win the pennant for three or four straight years, even though the team never makes it that high. It's like the AI is saying "OK, this is the year, I just know it!"

Just like evey publication known to man for the Cubs 8 of the last 10 years.

JMDurron · 01-19-2012, 07:22 PM

I enjoy the feature. It gives me a rough estimate of the relative strengths and weaknesses of particular teams/players, but not with enough precision to ruin the suspense. I think it tends to over-exaggerate the dominance of more talented teams, my more dominant rosters rarely win as many games as the predictions would indicate.

The closest prediction I ever saw was this one:

I ended up at 110-52.

As an aside, I would like to register my personal approval of the Endor Capitals' best pitcher.

01-17-2012, 06:03 AM	#1
Chicagofan76 Banned Join Date: Oct 2009 Location: Diamond, IL Posts: 6,339 Infractions: 2/2 (3)	Is Preseason Prediction Ever Accurate? Last season my EBL Preseason Prediction had my team winning 128 games. I won 105 of 144. This year they have me at 98 wins. see the rest below. Strategy and finances are set to 1976. 1. Who pays attention to this feature? 2. Have you ever seen 1 spot on? 3. Post yours. Attached Images

01-17-2012, 11:43 AM	#8
Jaxxvain All Star Reserve Join Date: May 2006 Posts: 531	I haven't seen it "spot on", which is good. I think it has a good amount of "fog of war". But I do wish it would do a better job with the "realism" of the prediction. If you notice it often lists the SPs as having 40+ starts and the hitters almost always playing 160+ games (and seemingly not ever getting subbed out). __________________ Jaxxvain aka Molson ~~~~~

01-17-2012, 11:53 AM	#9
joefromchicago Hall Of Famer Join Date: Jun 2011 Posts: 3,630	In the season I just finished in my dynasty, the AI accurately predicted the pennant winners in both leagues, and even got the runners-up correct. That's unusual, but not unprecedented. What I've found is that the AI will sometimes fixate on a team, predicting it to win the pennant for three or four straight years, even though the team never makes it that high. It's like the AI is saying "OK, this is the year, I just know it!" __________________ American-Ethnic (and Canadian) Namesets Historical Minor League Schedules 1870s City/Team Nickname Randomizers "It's Usually Sunny in Philadelphia" weather mod

01-17-2012, 06:14 AM	#2
hfield007 All Star Starter Join Date: Jul 2009 Posts: 1,651	I don't want it to accurately predict the championship which I'm glad it doesn't. I do check it each season. It does give a very accurate prediction I have found for me in guessing who the top 3 teams will be usually in each league. (with a few misses here and there which I like).

01-17-2012, 08:27 AM	#3
BusterKing Hall Of Famer Join Date: Mar 2002 Location: Ontario, Canada Posts: 2,369	Wow, with that kind of record it seems like your team is stacked! You actually enjoy the fact that you are so dominant over the other teams and lack of competition? Just curious, because I personally like to make the game harder for myself and impose a lot of house rules so that I don't take advantage of the weak AI.

01-17-2012, 08:31 AM	#4
mouse001 Major Leagues Join Date: Jul 2011 Posts: 343	I think there's enough fog of war in the preseason predictions. If you have a big enough league it's good at seeing whose the top 50%, whose the bottom 50%. With plenty of room for surprises of course

01-17-2012, 10:35 AM	#6
sc_superstar Major Leagues Join Date: Sep 2011 Posts: 407	LOL @ Markus

01-17-2012, 10:37 AM	#7
Bluenoser Hall Of Famer Join Date: Mar 2002 Location: In The Moment Posts: 13,683	1. Who pays attention to this feature? I do. 2. Have you ever seen 1 spot on? No, and I hope I never do, because then I'd believe there was a problem with them.

01-17-2012, 12:33 PM	#10
Biggio509 Hall Of Famer Join Date: Aug 2007 Posts: 2,027	I think this is the standard formulas that stats guys use to predict teams' records in the real world. It is possibly something like this. Pythagorean expectation - Wikipedia, the free encyclopedia A Pythagorean expectation could be set as a function of all players' past runs and runs allowed. The Pythagorean expectation rarely nails the wins correctly but it is used as a measuring stick to determine if a team is under or over performing. For instance, I have read some article claiming the Houston Astros were over achievers in the 1990's to 2005 because the expectation was lower than actual wins.

01-17-2012, 01:30 PM	#11
Erik All Star Reserve Join Date: May 2002 Posts: 505	Is there a way to pull this report up mid-season? I ask because I just started a new season and forgot to look at this report during Spring Training. Now when I go to Reports and try and pull it up it says it doesn't exist and I have to wait for next Spring. I thought I read somewhere this report would be available during the entire year so we could reference it? How would I go about pulling it up? Thanks.

01-17-2012, 04:06 PM	#14
MizzouRah Hall Of Famer Join Date: Dec 2001 Location: Troy, Mo Posts: 6,251	I think the AI is quite challenging.. year 9 of my franchise and I have yet to make it to the WS. Won the WS with every verison previous to v12. Like most, the report gives you a good idea of who will be good and who will be bad.. but like real life MLB preseason magazines, I don't want it 100% accurate either. Run it after ST and you won't have to bookmark it as it will be under the reports section.

01-19-2012, 07:22 PM	#20
JMDurron All Star Starter Join Date: Mar 2011 Posts: 1,262	I enjoy the feature. It gives me a rough estimate of the relative strengths and weaknesses of particular teams/players, but not with enough precision to ruin the suspense. I think it tends to over-exaggerate the dominance of more talented teams, my more dominant rosters rarely win as many games as the predictions would indicate. The closest prediction I ever saw was this one: I ended up at 110-52. As an aside, I would like to register my personal approval of the Endor Capitals' best pitcher.