Minimum/maximum performances possible

[gkhd11a]

[url]http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=341612[/url]

And While Maddux here gives up more walks and HR's than he did on his card I was able to pitch him 50% more than he was able on his card and only give up 30 hits more than he did with the 50% less innings. A very great performance.

[JOSEPHKENDALL]

It can be done at an $80M cap. It can't be done at the higher salary caps such as $140M and up.

[rburgh]

I created the Joe Strikeout card and put him in Fulton 78 in a live draft league. His final line - 29-11, 3.18, 44 GS, 359.1 IP, 235 H, 64 HR, 114 BB, 582 K. His card has the normal number of x-chart rolls.

Mathewson 05 was on the same team, he was 24-13, 6.40, with 95 HR allowed in 303 IP.

Then I changed the ballpark to Petco and ran it again. JS was 31-9, 2.44, 351 IP, 236 H, 50 HR, 88 BB, 557 K's; Mathewson was 20-13, 3.55, 329.2 IP, 278 H, 51 HR, 105 BB, 208 K's.

[macnole]

[quote:0e8a436b67="joekendall"]It can be done at an $80M cap. It can't be done at the higher salary caps such as $140M and up.[/quote:0e8a436b67]

Yes and I wouldnt expect it to.
The aggregate lineup is worth nominally 45M in an 80M league...possibly twice that in a 140M league...

The ~10M pitching card though is the same card.

[boyer14]

[quote:14c56c9472]http://fantasygames.sportingnews.com/stratomatic/team/team_other.html?user_id=341612

And While Maddux here gives up more walks and HR's than he did on his card I was able to pitch him 50% more than he was able on his card and only give up 30 hits more than he did with the 50% less innings. A very great performance.[/quote:14c56c9472]

Looks to me like Maddux gave up 106 more hits than his card for his 100 additional innings of work. What am I missing?

[Valen]

[quote:1ea769b6d0]Could not disagree more on "never get great performances" All you need to do is play the standard 80 million dollar league and you can achieve them all the time. Here is a very good performance from Pedro Martinez 44 walks in 253 innings vs card of 32 in 217 which transforms to 37 and gave up 21 home runs vs 17 in the which converts to 20 adjusted for the innings. On top of that used an 8 million dollar reliever.

[/quote:1ea769b6d0]
You disagree and then turn around and give an example that proves my point. You used a Martinez card that has not a single walk on it and still yielded 44 walks which is right in line with what I have shown as the minimim performance. Try that same thing with the 2010 Cliff Lee card.

While Pedro's card that year was outstanding and a great card the 37 walks in 217 innings was not historic and is thus somewhat reproducable in lower cap leagues. Again, try getting anywhere in the neighborhood of accurate with Cliff Lee regarding walks.

Once again for clarification. I am not saying the model is incapable of reproducing all pitching performances. I am not saying it is incapable of reproducing good pitching performances. What I am saying is it is incapable of reproducing extreme positive pitching performance and extreme negative offensive performances.

[Valen]

The owner's kid turned 21 today and his dad game him an MLB contract for his birthday. He gets to play every game no matter how bad he is.

Turns out he makes great contact and never strikes out. But all he can do is hit weak tailor made double play grounders. Thus ever roll in every one of his columns is a gb()A. He collected no hits and never walked. :(

How does he fare in the simulation :?:

In 595 AB he collects 67 hits including 17 doubles. He drives in 29 runs including 3 GWRBI.

Thus we illustrate that extremely bad hitting performances cannot be replicated by the game engine. There is a minimum number of hits and doubles that a hitter is going to get no matter how bad they are. Of course in online draft league strat nobody cares because nobody would draft him anyway.

[PotKettleBlack]

is vaguely saddened that we cannot accept the simple truth.

Strat uses baseball statistics, but is not baseball. It is a simulation that does, on the whole, a pretty good job of reproducing something like a baseball game that approximates the real thing when the context is correct. But, at no point, should we expect the simulation to be perfect. Especially given that the core engine of the game is a 50 year old card and dice game.

Consider the process of building a sim. You build a decision tree for each event. You then need to have each player in a batter/pitcher confrontation modify that tree. There will be conflicts between the batters modification to the process and the pitchers, just as there are peculiarities in baseball, like Enrique Wilson's inexplicable success against Pedro Martinez (.244 career hitter hit .440 against Pedro). Try and replicate that with a sim engine... any engine.

Just saying, when you understand that strat is a game based on baseball, but only delivering about 70-85% accuracy in the sim, you can relax and enjoy things more.

[Roosky]

I completely agree! Take the cards and game for what they are worth, it is why some cards are better deals than others. Personally my goal is to build a winning team and I really dont care how close the stats are to real life. I do like the info given from Valen though because it reveals some things that may be useful in building a winning team.

[PotKettleBlack]

The flaw in the balance of strat is that it puts the fielding (via X-rolls) entirely on the pitcher's tree. I'd do something a lot more advanced with a computer, that would use a physics simulation to figure ball trajectory and speed, and compare that against fielder speed, jump, and read. Thus a .400 in dead ball would be hard to replicate against a team like the 2011 TB Rays.