clear	
	clear matrix
	set matsize 800
	set mem 500m
	cd [your partition 
	use lanham_pitching.dta

drop if year <1900

	** back out hitter ABs
	*


	replace BAOpp = . if BAOpp == 9.99

	gen abf= H/BAOpp  




	

	// sum data per pitcher



	collapse (sum) W L G GS CG SHO SV H ER HR BB SO  IBB WP HBP BK BFP GF R  GIDP IPouts abf, by(playerID yearID teamID lgID)


	// IP

	gen IP=IPouts/3



	merge m:m playerID using  lahman names data
	drop _merge




drop if year <1900






	
	*** 2024 pitcher data ***

merge m:m bbrefID yearID using bbr 2024 pitcher data,update 
drop _merge
drop if yearID==.

drop if IP==.
	



drop if SO ==.



*** determine performance values as rates


gen k9=9*SO/IP
gen bb9=9*BB/IP
gen hr9=9*HR/IP
gen hr_pip=HR/IP
gen k_pip=SO/IP
gen bb_pip=BB/IP
gen hbp_pip=HBP/IP

gen rapg=9*R/IP


*** calculate fipr ***


 gen fipr=.
 
 foreach y of numlist 1900/2024 {
 	di "year `y'"
	regress rapg k_pip bb_pip hr_pip hbp_pip [iweight=IP] if year==`y'
	predict yhat 
replace fipr = yhat if year==`y'
drop yhat 
 }

 
 

** standardize seasopn fiprs ***

***********************************
* Population-Weighted Variance Code
***********************************

* 1. Summarize ip within each year
egen total_ip = sum(IP), by(yearID)

* 2. Compute weighted mean of fipr
generate double weighted_fipr_product = fipr * IP
egen double total_weighted_fipr = sum(weighted_fipr_product), by(yearID)
generate double weighted_mean_fipr = total_weighted_fipr / total_ip

* 3. Compute weighted sum of squared deviations
*    (fipr - weighted_mean_fipr)^2 * ip
generate double wssd = (fipr - weighted_mean_fipr)^2 * IP
egen double total_wssd = sum(wssd), by(yearID)

* 4. population variance: divide by total weight
generate double popvar_fipr = total_wssd / total_ip

* 5. Population weighted standard deviation
generate double pfiprd_fipr = sqrt(popvar_fipr)

* 6. [Optional] Clean up intermediate variiples
drop wssd weighted_fipr_product total_weighted_fipr


rename weighted_mean_fipr wvar_fipr
rename pfiprd_fipr wsd_fipr
drop   total_wssd total_ip






gen z_fipr = (fipr- wvar_fipr)/wsd_fipr





/// drop outliers

    
	
    // Compute quartiles and IQR
    summarize z_fipr, detail
    scalar iqr = r(p75) - r(p25)
    scalar lower_bound = r(p25) - 1.5 * iqr
    scalar upper_bound = r(p75) + 1.5 * iqr

    // Replace outliers with missing values
 //   replace z_fipr = . if z_fipr < lower_bound | z_fipr > upper_bound
	replace z_fipr=. if z_fipr<-3.25 | z_fipr >3.25
	zscore z_fipr
	replace z_fipr=z_z_fipr*-1
	
    
sum z_fipr,d

    
	
*** career fipr plus ***

// weight each pitcher's season fipr by season IP / career IP 

* 1. Summarize ip within each year
egen total_ip = sum(IP), by(playerID)

* 2. Compute weighted mean of z_fipr
generate double weighted_z_fipr_product = z_fipr * IP
egen double total_weighted_z_fipr = sum(weighted_z_fipr_product), by(playerID)
generate double weighted_mean_z_fipr = total_weighted_z_fipr / total_ip



* 2. Compute weighted mean of fipr
generate double weighted_fipr_product = fipr * IP
egen double total_weighted_fipr = sum(weighted_fipr_product), by(playerID)
generate double weighted_mean_fipr = total_weighted_fipr / total_ip

* 3. Compute weighted sum of squared deviations
*    (z_fipr - weighted_mean_z_fipr)^2 * ip
generate double wssd = (z_fipr - weighted_mean_z_fipr)^2 * IP
egen double total_wssd = sum(wssd), by(playerID)


rename weighted_mean_z_fipr z_fipr_avg


drop if total_ip < 1000


collapse fipr z_fipr_avg total_ip, by(nameLast nameFirst )

gsort -z_fipr_avg



sum z_fipr_avg,d


** normalize on mean 100 plus scale

gen fipr_plus_norm = 100 * (z_fipr_avg - r(min)) / (r(max) - r(min))





sum fipr_plus_norm,d
gen fipr_plus = (fipr_plus_norm/r(mean))*100



egen rank_fipr_plus_desc = rank(-fipr_plus)


sort rank

format fipr %9.2f
format total_ip %9.0f
format fipr_plus %9.0f

rename fipr career_fipr
rename total_ip IP
export excel rank nameFirst nameLast  IP career_fipr fipr_plus  using "standardized_career_fipr.xls", firstrow(variables)   replace