I thought I'd start a quasi-blog here about what's happening with development for HD. I'll try to post something here somewhat regularly when there is significant active development happening. I'll keep this thread locked to keep things clean, but feel free to start a new thread for discussion about any posts made here.

As I've mentioned before in various spots, the next release will be huge, including several key areas of the game. The SimEngine is being translated to a new programming language and so it's pretty much a rewrite. Some of the logic will remain the same, some will be tweaked, and some will change more significantly. It's a little early to say what will change and what won't, but our focus area is emphasizing individual matchups more. This process will most importantly give me a complete understanding of the SimEngine and allow for easier improvements in the future.

One change that I want to mention is that durabilty will play a bigger part in injuries in the next SimEngine, so plan accordingly. You may want to start looking harder at durability. We're not planning on drastically changing the frequency of injuries, but we do want durability to be meaningful.

Another key target is recruit creation. We're changing the methodology there to allow for more variability in player quality, making truly elite players stand out more. This will inevitably involve a lowering of ratings across the board in order to keep the 1 to 100 scale. So there may be some growing pains at first as new players may not be quite as good as in the past, but in the long run it will make things better. Another result of the changes will be more unique players, where they're not tied so much to a traditional mold. So you should see more perimeter big men for example.

We're also revamping the top 25 ranking system along with the tournament selection/seeding process. I know there has been some disconnect between the two and we're trying to bring them more in line.

It's important to let you guys know that we are working to improve the game and we're targeting the areas that generate the most feedback.

Thanks.

I should also mention a timeline: we're probably looking at 2-3 months before this release is ready as we want to thoroughly test it beforehand. We'll be asking for testers in a month or so probably.

Things are still chugging along here. Lots of work happening in the language translation process that isn't useful to pass along to you guys. One thing that is of interest is that we're not going to do a full translation of the entire game logic at once, so after the next release game simulations, recruit creation, rankings, and probably some tournament seeding things will be separate from the rest.

In the newly translated processes, the timing of when things run may change a little. The cycles will still be at the same times, but most likely the new processes will run on a world by world basis. In other words, Naismith will complete all of its tasks before any other world processes and so forth down the line. In the current process, it's the other way around, where games run for all worlds, then recruiting is processed, etc. I don't think it will drastically change things since there are a small number of worlds anyway, but I wanted to give a heads up.

Not much has happened since the last post as I've been pulled off for some more urgent work, but I'm ready to refocus now and should have nearly all of my time for the near future to devote to this update. I've responded to people asking to be testers to watch the HD forum for us to ask for people. That will still happen, but probably not for another month or so. There's no point in collecting testers before there's anything to test.

I appreciate the patience everyone has shown and I'm working hard to reward it.

Quick note about rankings: the plan at this point is to have them update after each game instead of every two games. It takes away that bit of realism, but it makes things simpler on our end and probably makes you guys happier too.

I may have mentioned this already, but our plan for the rankings is that there will be two components to them. One will be some measure of the talent of a team based on ratings. The other will be performance in games. Early in the season the talent portion will be weighted heavier than performance, but as the season progresses, that balance will shift to performance being the dominant piece.

Working on recruit creation now. It's shaping up and we're getting close to where we want to be. Overall it probably won't look that much different, but there will be more of a bunching toward the average with fewer extremes. But those extremes may be a bit better than what you get now.

Also, I'm tweaking the selection of favorite school a bit. It will be spread out a little more than it is currently. Basically, if a player wants to stay close to home, his favorite will be a school pretty close to him and vice versa if the player wants to be far from home. If he has no preference it could be a nearby school or a nationally prominent school based on prestige. But the latter group will include more schools than it does currently. Also, a recruit's favorite should now be a school that he will actually be interested in attending based on his talent. Overall, maybe not quite as realistic but should work better within the context of the game.

We're still ironing out the details, but the player rate of improvement will be tweaked for the next release. The change earlier this year helped quite a bit, but I think it needs another smaller adjustment to slow it down.

I know one of the biggest complaints about potential has been that your players max out pretty quickly and then there is no reason to set up a practice plan any more. So hopefully we can address that with a combination of slowing the improvement rate and increasing the potential a bit for most players. That may also lead to slightly lower starting ratings, but I don't think they'll be significantly lower.

I want to mention that, for the new SimEngine, Low Post will essentially mean "ability to score in the paint area". In the current SimEngine Low Post is pretty much insignificant for guards, but that will not be the case for the next release. There will even be guards who are stronger in Low Post than they are in Perimeter. On the flip side, you'll see big guys who are stronger in Perimeter than Low Post. I just wanted to get that bit of news out since I know some coaches ignore Low Post when recruiting guards.

On another note, the plan is for lower division players to have more upside/potential than they currently do. I think that will make a lot of the DII/DIII coaches happy.

We've done some analysis of tournament seeding in several recent NCAA seasons and determined that the factors that significantly influence a team's seed are RPI, SOS rank, and top 25 ranking. We actually rank all teams during the top 25 ranking process, so there is no limitation to using that.

So the plan to start with is to weight those 3 factors to determine both who gets into the tournament and where they're seeded.

It's possible there may need to be some other factors considered, but the simpler it is the easier it is to understand why one team gets a bid over another. The new ranking system should fall more in line with RPI anyway, so hopefully there won't be cases of ranked teams with low RPIs being out or seeded low.

Also, I'm going to be wait until after Christmas to open up the test world. I'm using a copy of Knight and I'd like to copy down the current data right after the end of the next recruiting cycle for testing purposes. That will give coaches who are in Knight a base knowledge of the team they'll be testing with to better compare once the new engine is in use for testing.

Again, I really do appreciate everyone's patience during this process. Obviously the original intent was to get this release out the door much earlier, but there have been a few unforeseen delays. I'm still very excited about this though and I hope you all are too.