Monday, February 10, 2014

Blood Bowl: Preseason Goals

One of the FumBBL leagues I'm in runs a 4 week preseason before the 14 week regular season. The league has a salary cap which applies before the preseason so the preseason is a way to let teams build back up a little after potentially having made some brutal cuts. (There's one dark elf team currently running with 1 reroll!) They also build the divisions for the season based on the standings of the swiss preseason in a snake format. Last season my division had the 3, 14, 19, and 30th teams from the preseason standings for example.

I've been thinking about what I should be trying to accomplish in the preseason. My team doesn't have any glaring weaknesses from the salary cap but I do have a couple of no-skill players who could use a cheap level. I've been thinking about buying a minotaur to level him in the preseason but since I had to light most of my cash on fire doing so would risk not playing with some positionals if I actually start having people die. Should I be trying to win the preseason? Should I be trying to kill my opposition? Should I just be trying to keep my own guys alive and screw winning?

Last season I went the protect my own guys route by inducing extra apothecaries whenever I had inducement money to spend. The idea being that while I like winning I wanted to be able to win in the regular season and that meant keeping people alive as I built up the money and skills to have a reasonable team. Apothecaries are not very good at winning games, certainly not compared to picking up star players, but the star player can steal the MVP experience and the apothecary can save someone's career. Not guaranteed to, of course, but Blood Bowl is a game of small edges and those felt like good edges to pick up.

I still tried to win, of course, since scoring is worth experience and winning is worth extra money. I went 1-2-1 which seemed like a decent result and made me #14. I also tried to hurt people because that's how Blood Bowl works. My last preseason game included 6 serious injuries on my opponent which set him way, way back. But doing so wasn't really to my advantage, especially since so many of the injuries weren't even worth experience (failed leaps mostly)! I did get to play that team in the season but I wasn't one of the first couple who got to take real advantage of the damage I did. So while doing damage is fun it isn't really profitable for the season.

Winning does have a tangible benefit in a salary cap boost for the next season for the top three teams. I didn't think anything of that last season but ended up well over the cap due to the great luck I had avoiding injuries all season long. My team this season is already at the cap for next season so any cap bonuses I could pick up would seem to be pretty relevant.

How about division seeding? Is that relevant? Winning the preseason would mean getting into a division with the teams that came 16, 17, and 32nd. If everyone is playing all out to win then getting two games against the worst team and having no one in the top 15 sounds awesome! But maybe with new teams every season and the randomness of swiss pairings in a 32 team pod the seedings won't actually have much meaning? Well... What happened last season? Did the preseason standings have much predictive power towards the final standings?

I plugged the preseason ranking from last season into Excel along with the points earned during the season by those teams. The two sets of data were moderately negatively correlated with an r value of -.594. So certainly not a guarantee but it does seem like it would be a decent predictor. All the teams that had really bad regular seasons also had a bad preseason. It looks like the teams in the 12-21 range all did better than would be expected and the teams in the 5-11 range did worse. I wonder if that has anything to do with the snake factor? Or if this is just a really small sample size? I couldn't find the final standings from previous seasons because the images look to have been removed or something so I couldn't check for prior results. I tried replacing preseason ranking with preseason points (avoiding tiebreakers) and r became .527. It still shows that the preseason and the regular season are correlated, but not a super lot. It's almost like Blood Bowl teams can change significantly over time!

Anyway, it's a big enough thing that at least getting into the top 8 seems like it could be a good thing? And getting all the way to first is probably really good. I do want the cap bonus and I'm becoming convinced that division seeding is at least a little relevant. But I don't think going super crazy to win is going to make sense. I'm still going to save my apothecary for a serious injury I think. Or maybe I won't! I am a bit of a wild man...

I've already played one game this preseason and my opponent induced up a claw star player and used him to blitz almost every turn. It seemed like a good way to damage my team and a questionable way to skill up his own team. But he did get 14 SPP on the rest of his team so I guess it worked out pretty well. I don't think I would have taken the risk of 2 star players getting the MVP but he didn't get dinged. And I actually ended up permanently injuring one of them, so maybe having a bomber as a lightning rod made sense. I won 2-0, which is good for me, but I suffered a stat loss which is bad for me. Well, quasi-bad. Now one of my dwarves has only 1 agility. It probably won't matter and certainly isn't worth retiring a 28 SPP dude over but I wish it hadn't happened. Poor Mrs White.

No comments: