Project 1 - Chess Player Rating
sanjive
September 21, 2015
Project Goal: Extract chess Player Name, State Name, Pre Rating and New Rating as an average from the Win, Loss and Draw points of the opponent and convert the new extracted dataset to a csv file for easy upload to any SQL DB.
Initiated project with the tournamentinfo.txt upload to Github for easy access to everyone and fetching the data universally. Once the file data is accessible, moved forward with the first step as: - Extraction of Data: Extraction of data started up with analyzing the data pattern which looks like ithas two different pattern of data as dataLine1 and dataLine2. - Once we have all the data in dataLine1 and dataLine2, started up extraction of playerName, stateName and totalPoints - Now the challenge is to get the Pre rating data without “P” suffix, so that this can be used to calculate the New Rating. This was achieved by replacing the P[0-9] with “”, now we are left with the string data set, which was converted to numeric for further calculation
library(stringr)
tournament_info <- readLines("https://raw.githubusercontent.com/ksanju0/IS607/master/tournamentinfo.txt")## Warning in readLines("https://raw.githubusercontent.com/ksanju0/IS607/
## master/tournamentinfo.txt"): incomplete final line found on 'https://
## raw.githubusercontent.com/ksanju0/IS607/master/tournamentinfo.txt'dataLine1 = unlist(str_extract_all(tournament_info,"^[[:blank:]]+\\d{1,2}.+"))
dataLine2=unlist(str_extract_all(tournament_info,"^[[:blank:]]+[A-Z]{2}.+"))
playerName=unlist(str_extract_all(dataLine1,"(\\b[[:upper:]-]+\\b\\s)+(\\b[[:upper:]-]+\\b){1}"))
stateName=unlist(str_extract_all(dataLine2,"[[:upper:]]{2}" ))
totalPoints=as.numeric(unlist(str_extract_all(dataLine1,"\\d(.)\\d")))
cleanpreRatingData<-str_replace_all(dataLine2,pattern="[P]\\d{1,}"," ")
line21 <- str_extract_all(cleanpreRatingData,"([R(:)][[:blank:]]+\\d{3,}+)")
preRating<-as.numeric(str_extract_all(line21,"\\d{3,}"))Now next steps is to find the opponents numbers with whom each of the players had either Win, Loss or Draw
Opponents1<- str_extract_all(dataLine1,"[WLD][[:blank:]]+\\d{1,2}")
OpponentsData<-str_extract_all(Opponents1,"\\d{1,2}")
opponents <- lapply(OpponentsData, as.numeric)Now we have all the data required to calculate the new Rating, but before that we need to bind all the data together in a data frame (PlayerDF) for easy access and reference while doing the calculation for ave_newRate as a function. This function will calculate the average rating of the player based on Win, Loss or Draw and return the complete dataset as an object to newRate. Once we have the dataSet, convert strings to numeric and format it to match with the Pre rating. Finally we cna bind this newRate also with the data frame (PlayerDF) and write it in any format for future use, here we have write it in csv format.
PlayerDF<- data.frame(playerName,stateName,totalPoints,preRating)
ave_newRate <- function(x){
Newrating<-0
totOpponents<-length(x)
for (i in x){
Newrating<-Newrating+PlayerDF[i,"preRating"]}
return(Newrating/totOpponents)
}
newRate <- unlist(lapply(opponents, ave_newRate))
newRate <- as.numeric(sprintf("%1.0f",newRate))
PlayerDF<- data.frame(playerName,stateName,totalPoints,preRating,newRate)
PlayerDF## playerName stateName totalPoints preRating newRate
## 1 GARY HUA ON 6.0 1794 1605
## 2 DAKSHESH DARURI MI 6.0 1553 1469
## 3 ADITYA BAJAJ MI 6.0 1384 1564
## 4 PATRICK H SCHILLING MI 5.5 1716 1574
## 5 HANSHI ZUO MI 5.5 1655 1501
## 6 HANSEN SONG OH 5.0 1686 1519
## 7 GARY DEE SWATHELL MI 5.0 1649 1372
## 8 EZEKIEL HOUGHTON MI 5.0 1641 1468
## 9 STEFANO LEE ON 5.0 1411 1523
## 10 ANVIT RAO MI 5.0 1365 1554
## 11 CAMERON WILLIAM MC LEMAN MI 4.5 1712 1468
## 12 KENNETH J TACK MI 4.5 1663 1506
## 13 TORRANCE HENRY JR MI 4.5 1666 1498
## 14 BRADLEY SHAW MI 4.5 1610 1515
## 15 ZACHARY JAMES HOUGHTON MI 4.5 1220 1484
## 16 MIKE NIKITIN MI 4.0 1604 1386
## 17 RONALD GRZEGORCZYK MI 4.0 1629 1499
## 18 DAVID SUNDEEN MI 4.0 1600 1480
## 19 DIPANKAR ROY MI 4.0 1564 1426
## 20 JASON ZHENG MI 4.0 1595 1411
## 21 DINH DANG BUI ON 4.0 1563 1470
## 22 EUGENE L MCCLURE MI 4.0 1555 1300
## 23 ALAN BUI ON 4.0 1363 1214
## 24 MICHAEL R ALDRICH MI 4.0 1229 1357
## 25 LOREN SCHWIEBERT MI 3.5 1745 1363
## 26 MAX ZHU ON 3.5 1579 1507
## 27 GAURAV GIDWANI MI 3.5 1552 1222
## 28 SOFIA ADINA STANESCU-BELLU MI 3.5 1507 1522
## 29 CHIEDOZIE OKORIE MI 3.5 1602 1314
## 30 GEORGE AVERY JONES ON 3.5 1522 1144
## 31 RISHI SHETTY MI 3.5 1494 1260
## 32 JOSHUA PHILIP MATHEWS ON 3.5 1441 1379
## 33 JADE GE MI 3.5 1449 1277
## 34 MICHAEL JEFFERY THOMAS MI 3.5 1399 1375
## 35 JOSHUA DAVID LEE MI 3.5 1438 1150
## 36 SIDDHARTH JHA MI 3.5 1355 1388
## 37 AMIYATOSH PWNANANDAM MI 3.5 980 1385
## 38 BRIAN LIU MI 3.0 1423 1539
## 39 JOEL R HENDON MI 3.0 1436 1430
## 40 FOREST ZHANG MI 3.0 1348 1391
## 41 KYLE WILLIAM MURPHY MI 3.0 1403 1248
## 42 JARED GE MI 3.0 1332 1150
## 43 ROBERT GLEN VASEY MI 3.0 1283 1107
## 44 JUSTIN D SCHILLING MI 3.0 1199 1327
## 45 DEREK YAN MI 3.0 1242 1152
## 46 JACOB ALEXANDER LAVALLEY MI 3.0 377 1358
## 47 ERIC WRIGHT MI 2.5 1362 1392
## 48 DANIEL KHAIN MI 2.5 1382 1356
## 49 MICHAEL J MARTIN MI 2.5 1291 1286
## 50 SHIVAM JHA MI 2.5 1056 1296
## 51 TEJAS AYYAGARI MI 2.5 1011 1356
## 52 ETHAN GUO MI 2.5 935 1495
## 53 JOSE C YBARRA MI 2.0 1393 1345
## 54 LARRY HODGE MI 2.0 1270 1206
## 55 ALEX KONG MI 2.0 1186 1406
## 56 MARISA RICCI MI 2.0 1153 1414
## 57 MICHAEL LU MI 2.0 1092 1363
## 58 VIRAJ MOHILE MI 2.0 917 1391
## 59 SEAN M MC CORMICK MI 2.0 853 1319
## 60 JULIA SHEN MI 1.5 967 1330
## 61 JEZZEL FARKAS ON 1.5 955 1327
## 62 ASHWIN BALAJI MI 1.0 1530 1186
## 63 THOMAS JOSEPH HOSMER MI 1.0 1175 1350
## 64 BEN LI MI 1.0 1163 1263write.csv(PlayerDF, file ="playerDF.csv")Further analysis to find out the data distribution pattern for new Rating as shown below using normal density plot and further confirmation to teh normal disdribution is done using normal Q-Q Plot
par(mfrow=c(1,2))
newRateMean <- mean(PlayerDF$newRate)
newRateSD <- sd(PlayerDF$newRate)
hist(PlayerDF$newRate,probability=TRUE)
x <- 800:1900
y <- dnorm(x = x, mean = newRateMean, sd = newRateSD)
lines(x = x, y = y, col = "blue")
qqnorm(PlayerDF$newRate)
qqline(PlayerDF$newRate)
summary(PlayerDF$newRate)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 1107 1310 1382 1379 1481 1605Conclusion: Histogram shows the unimodal distribution with mean almost equal to median as 1379 emphasizing the normal distribution pattern. This was reinforced by the normal QQ plot which also suggest as most of the data is around the mean/median and dense at 1310 to 1379. it also has outliers as 1605 and few more data point beyond 1500.