alexfall862

Hello Coaches! Welcome to the first Dev Diary for 2022 on the new recruiting system for this coming season. Recruiting Dev Diaries for 2022 Dev Diary 54: Recruiting Affinities Refresh | Close to Home Dev Diary 55: Recruiting Affinities Refresh | Other Affinities Dev Diary 56: Recruiting Efficiency Score Dev Diary 57: Recruiting on the Interface With this diary, I want to take a moment to talk about the changes to close to home assignment for the next season. The goal for this offseason was to look at all affinity assignments and make sure teams were getting bonuses that made sense and that the system overall was reflecting something close to reality. This Dev Diary is for the Close to Home affinity, which affects around 60% of all recruits in a given class. Close to Home Score In order to compare schools' situations as close to objectively as possible, I developed a score to reflect the strength of a team's close to home assignments. There's a little bit of math behind it, but the gist of the score is that the higher the number, the stronger the close to home advantage that team should have. When comparing schools, first let's look at total number of players each school could recruit with a close to home bonus. This is the total number of players in CtH states for schools, multiplied by 60% (the percentage of recruits generated with close to home affinity): This is certainly one measure to see which schools have an advantage with close to home as an affinity, but one thing it doesn't capture is the degree to which multiple teams are competing for the same resources. After all, is Hawaii really the worst place to try and recruit players who want to stay close to home? Is West Virginia really that strongly positioned? So instead of looking at total recruits, let's use the Close to Home Score. This score does a good job of reflecting close to home advantages both in the sim and in real life. You can see all the rocky mountain schools tearing each other apart at the bottom due both to having few recruits in their regions, but also due to so many schools being in the same region. Instead of being ranked as the weakest school for close to home advantages, Hawaii is now in the upper middle of the rankings. This makes some sense, because while they may only have 25 recruits, there's no other competition that can argue a recruit could stay close to home without playing for Hawaii. We've spent some considerable time talking about the Close to Home Score, and I want to emphasize that this isn't a measure we tried to fit our schools to, but a score that lets use know whether our existing assignments pass the 'smell test'. The actual work of assignment and reassignment of close to home regions was done with quite a bit of hand-checking of regions and whether a school would plausibly be able to make the argument to a recruit that they'd be able to stay closer to home than their other options. In general, as far as straight geography, schools in the middle of the country have a geographically larger orbit in close to home regions and the coasts have a much tighter window to argue for close to home. My sense is that this reflects reality, as communities on the coast tend to be denser on the coasts and sparse in the plains, particularly the northern plains. 2022 Close to Home Changes All of the above is a preamble to this - the changes that have been made for 2022 to Close to Home affinity assignments! You can look at the 2022 specific state assignments here and see an interactive map here. (One thing the map can't do is divide between the sub-state regions we have for California, Florida, and Texas. Copy the drive document to avoid two users trying to select things at the same time) Wrap Up I asked some users to help out with this job, and special thanks need to go to @PoopyRhinoPickle @subsequent @TuscanSota @Rocketcan @tsweezy for their help and especially to @PoopyRhinoPickle as this dev diary would probably have been posted a month from now without his help compiling all of the changes.

Ah dang, no poison pills?

Nebraska alexfall862#6505

Nebraska alexfall862#6505

Hello Coaches! A Dev Diary I'm sure you're all interested in! It's time to talk 'croots! General overview of recruiting can be found here: https://www.simfba.com/index.php?threads/recruiting-tutorial.2353/ Nuts and Bolts Recruiting will be a weekly activity, with actions taking place whenever you'd like during the week, but progressions occurring Tuesday nights for a Wednesday to Wednesday recruiting week. The recruiting season will last 20 weeks. The first week of recruiting will take place from the Wednesday prior to Week 1 to the Wednesday just prior to Week 2. This means that the recruiting season should wrap up Wednesday following the last bowl games. Recruiting is done by assigning recruiting 'points' out of your weekly recruiting budget to a player. When a player has collected enough points, they will pass a threshold and make a 'commitment decision' to eligible schools. In order to be an 'eligible school' when the commitment is assessed, you must have a standing scholarship offer to the player and be in their leading schools list. The purpose of this dev diary is to dive into the commitment decision and how the leading schools list in generated. Tutorial about the user-side of recruiting is here: Leading Schools The leading schools function lists the recruiting points leader and schools within striking distance of the leading school. This math is set at 50% of the leader's amount of points. Towards the time players will start making recruiting decisions, being listed here means that you are (roughly) within a week's point of becoming the points leader. When a commitment decision is calculated, the initial list of schools eligible to land the recruit are generated from the leading school, then any schools without a scholarship offer are removed. This means that you can be listed in leading schools and have 0% chance to land a player if you didn't offer them, but also means that you can offer a player and spend points on them and also have 0% chance to land them because you're not in their leading schools list. This is meant to be analogous to real-life recruits' Top X lists. Commitment Decision A player will make a decision to commit to a school when they have passed a certain threshold of total points. That threshold is not static throughout the year, it is changed by two inputs: the recruiting week and the total number of teams recruiting the player. The longer the year goes on, the lower the threshold required becomes, with Week 20 culminating in the every player left making a decision to essentially anyone who recruited them at all. The number of teams raises the amount of points needed proportionally - meaning that the amount of teams pursuing a player don't impact their 'rush' to commit, but in a lot of cases cause a lengthening of their recruiting process. What decides how quickly a player commits is the average point total per team. If a recruit has an average of 20 points per team per week recruiting them, they'll make a decision when we advance into Week 4. If a recruit has an average of 10 points per team per week recruiting them, they'll make a decision when we advance into Week 6. If a recruit has an average of 1 point per team per week recruiting them, they'll likely wait until advance into Week 19. Once the player passes the recruiting threshold, the teams that are eligible (on leading schools list + scholarship offer from school) have their point added up. The points become their odds of landing a player. For example, if School A has 40 points invested into a player, and School B has 60 points, then School A will have a 40% chance of landing the player, and School B will have a 60% chance of landing the player. If School A has 30 points invested into a player, School B has 35, and School C has 40 but didn't offer a scholarship, then School A has 30/65 odds, and School B will have 35/65 odds of landing that player. School C's points might have triggered the recruit's commitment decision, but since they didn't offer a scholarship, their odds of landing that player are 0%. The goal here is to allow for a more realistic experience when recruiting players, where min/maxing to edge out other schools by 1 point doesn't make a large difference in outcomes, but managing your targets and point spending still rewards coaches. Affinity Bonuses Aside from players directing points from their budget into recruits, the other way to gain points is to capture what are called 'affinity bonuses'. Some recruits have 'affinities' assigned to them. All but one affinity is static, meaning any school with that affinity is eligible to get the bonus for any recruiting with that affinity. The affinities that fall into that category are: - Academics - Climate - Service - Religion - Large Crowds - Small School - Frontrunner Right now, one of them is recruit/school context dependent, meaning that the school needs to match certain criteria instead of also holding that affinity. In the future,the list of context dependent affinities will likely grow, but for now that one affinity is 'close to home'. If the school is in a location in the specific recruit's home zone, AND they have the 'close to home' affinity, they will be eligible for the bonus. The bonus itself is an additional 25 points for the 5th, 10th, 15th, and 20th consecutive weeks you invest at least 1 point in a player. These affinity bonuses are only given to schools that are 'actively recruiting' that player. What that means is that to get the 5th week affinity bonus, you need to invest a minimum of 1 point in (for example) weeks 1, 2, 3, 4, and 5. You could also get the 5th week bonus using weeks 6-10 or 12-16. The 10th week 25 point bonus requires at least 1 point in (for example) weeks 6, 7, 8, 9, 10, 11, 12, 13, 14, and 15. Weeks 15 and 20 bonuses follow the same pattern. I hope this gives an in-depth review of the recruiting system's machinery. As we go from one test-run to a full sim, I'm sure there will be edits to processes, and we'll keep folks in the loop as changes are made. - alexfall862

alexfall862 and Nebraska will be participating this season. Discord is @alexfall862#$6505

FIRE LARRY SCOTT [uSER=1]@SUBSEQUENT[/uSER] But also, woo! Getting closer to kickoff!

I'm super bummed I'm not in a position to help anymore, but this is just the coolest shit.

Fun fact: I did the design for their charity website! Bunch of really awesome folks. Happy to see it still going, even if we haven't had the trophy for a while...

@Minnow Only if the bits are broken off of your back!

@Nittany Boiler @subsequent My understanding is that the player database is the same for recruits/teams - so they retain all the 'tags' they had during recruiting. I might be wrong on that since I've been out of pocket for a while. If that's still correct, maybe something for the non-immediate future is tie transfers to recruiting affinities (when we get to transfers)? If a recruit has an affinity for immediate playing time (or maybe 'Going Pro' as well), but ends up redshirted or has <50 downs played in a season (with no injury) they have a high chance to transfer. It could be stored as a 'frustration threshold' and players accumulate and lose points at the end of each season for things like playing time, overall record, or a new head coach. If they get over a certain point total, the database flags them for transfer season. Since playing time is something players of any skill could end up with as an affinity due to the procedural generation, you'd see players transferring off the bench and potentially to 'little sister' schools where they can hit the top of the depth chart or even leaving smaller schools if a team is min/maxing recruiting in some strange way and has recruited 5+ QBs or something. Long-term, I could envision a way of handling transfers that took the coaches' history into affect, where they had a higher chance of transfers from their previous school (if player was on the team then). The nuts and bolts are mostly there to create some really fun emergent narratives, but it's shouldn't be a priority atm obviously.

This is really great! Nebraska v OU when? @HAFFnHAFF

1. What is your username? alexfall862 2. Which team do you want to apply for? Nebraska 3. What are 3 backup teams in descending order of preference? ... 4. How much time can you devote to the community? The more people that are active, the more fun the sim as a whole is with all of the fun interactions that can take place between users, whether it be in the Shoutbox, or through various forms of media. 2-3 hours a week. 5. How did you find out or hear about SFA? Helped put some stuff together with Rocketcan and Swipet most recently, though tons of awesome folks are helping out. Especially Kirby - great work on the website forum! 6. Do you know anybody from SFA? List their usernames if applicable. Lots of 'em. Most everyone in the Shitbox. 7. Tell us a little about yourselves to serve as a short introduction to the community. Working on the Sim with Rocketcan, was the winningest coach in CFBHC history at Wyoming (sub .500 record).