Did you catch any of the Wales-England game? One of the English backs tried to dink a kick into the corner of the in-goal area and ended up chipping it into the crowd, was hilarious.
Working on Pick and Goes at the moment. Here's the theory I'm working on. Pick and Goes will mainly be deployed by the attacking team within 5m of the opposition try line. The most likely players to be used for the pick and go are forwards. However, I don't think it'd be unreasonable to use a high quality back rather than a poor quality forward - especially a crash ball type centre. Hence, the player to be used at the pick and go will be =Random Factor + Strength *2 + Stamina * 0.5 + Crash Ball * 0.5 + Speed * 0.5 This calculation is purely to determine who will attempt the crash ball. The theory being: Strength - you want your strongest player to make the attempt. This will bias selection in favour of forwards, and particularly to top quality props / flankers. Stamina - you've got to get to the ruck in the first place in order to attempt a pick and go. The more fitness the player has, the more likely to be Johnny on the Spot Crash Ball - more likely to gain yardage at a pick and go if you've got decent crash ball skills. Speed - similar to stamina Thoughts?
Deferred to Season Six - Cutting lineouts down to 3 or 5 man lines (LOW priority) - Have line outs separated from line out mauls (to allow the maul resulting from the lineout to have be more specifically dealt with) (MED priority) - Quick Throws (LOW priority) - Wind (LOW priority) - Pitch conditions (LOW priority) - Marks (MED priority) Season Five - Split Pick 'n' Gos from other open play moves (HIGHEST priority) - Change Penalty kicks at goal calculation. Will make distance that kick can be made from dependent on players place kicking skill. Also, now the sim keeps a tally of the scoreline, I should make Penalty kicks dependent on the scoreline as well in the same way that I've made Drop Goals dependent on the scoreline. (HIGHEST priority) - Change distance of kicks in open play to be more dependent on kicking skill - Enhance commentary (LOW priority) - Engine change - rebalance ruck/open play split (potentially change tackle calculation?) (HIGH Priority) - Every player to have some limited ability to make successful place kicks. (HIGH priority) - Add position played in to the stat output - Separate Free Kicks out from Kicks in Open Play - Change Poor Binding calcs so that 2nd rowers can't be pinged for this so easily. Reduce likelihood of feed errors from top quality scrum halves - Devise a way to sim added time without having to update the stats manually. (HIGH priority) - More stats - (HIGH priority) 1. Rucks contested 2. Line out throws straight 3. Line out throws not straight 4. Maul metres 5. Turnovers Conceded by Ball Carrier 6. Number of runs attempted 7. Proper missed tackle stat I think we're done here. I need to make sure Pick and Go stats are counted, but other than tackles/missed tackles/metres gained - what more do we need? Ideally, I wouldn't have to monitor Pick and Goes seperately (i.e. number of PnGs/metres gained from PnGs etc.) but if there's massive demand than I can add them. Or add them at a later date.
In the end, I've added a Number of Pick and Goes by Player stat. I probably have to adjust the player calc though. At the moment, about a quarter - a third of all pick and goes are attempted by backs. Top quality crash ball backs are skewing things a bit, so I think I'll adjust so that the impact of crash ball is reduced, or maybe add in a need for technique as well.
You need a different formula for each position if you want to compare all 15 guys against each other: Hooker Props 2nd Row Back Row etc The kicker in the team should have his own calculation. You will find that it will be impossible to get the coefficiants set to just magically give everyone comparable scores. So after experimenting on past games - you will have to weight each score - ie backs get multiplied by 0.85 - props *1.12
No, I wouldn't have time. That doesn't stop someone volunteering to do that for the Seasonal Rugby Gazette though.
I have no idea how to work out coefficients based on the outputs to date as there's multiple variables in play - you can't just compare two different players in two different positions with similar postcounts, sim a number of games and work out an expected difference between the two. What I'm trying to replicate is the SuperCoach style of player scoring. Some players and positions are always going to score more points than others, but that's the nature of a SuperCoach style of scoring.
Added a nice little enhancement which will significantly reduce the likelihood of kicking the ball when you're on the attack and within 20m of the opposition try line. Does the 20m sound reasonable? I think it should probably be increased to 30m So, basically (depending on your tactics & the weather conditions) on the occasions where the next play is chosen at random, there was between 1/40 and 1/5 chance of going for a kick. There's now a second check which will increase the probabilities to (1/40)*(1/40) to (8/40)*(8/40) i.e. (1/1600) to (1/25). Should hopefully reduce the chances of making one of those silly kicks from inside the 22 whilst on the attack.
Yeah, and the rankings will work out the mean and standard deviation for each position based on that, so the comparisons there at least will be completely relative.
