DATA607 Final Presentation - NHL Salary Cap Analysis

John Grando
December 20th, 2017

NHL Salary Cap Structure

plot of chunk unnamed-chunk-5

Player Salaries

plot of chunk unnamed-chunk-6

Player Salaries

plot of chunk unnamed-chunk-7

Curve Fit - Forwards


Call:
glm(formula = CAP.HIT.. ~ CF + Goals + Total.Assists + Shots + 
    Total.Penalties + Minor + Takeaways + Hits.Taken + CF_pp + 
    CA_pp + SCA_pp + HDCF_pp + LDCA_pp + Goals_pp + Total.Assists_pp + 
    Shots_pp + iSCF_pp + iHDCF_pp + SCA_pk + HDCF_pk + HDCA_pk + 
    Total.Assists_pk + iCF_pk + iSCF_pk + Rebounds.Created_pk + 
    Penalties.Drawn_pk, family = Gamma(link = "log"), data = f_train_df)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-1.2470  -0.2638  -0.0255   0.2118   0.8513  

Coefficients:
                    Estimate Std. Error t value Pr(>|t|)    
(Intercept)          14.2087     0.1867  76.103  < 2e-16 ***
CF                    0.4760     0.2324   2.048 0.041183 *  
Goals               -10.0489     4.3190  -2.327 0.020449 *  
Total.Assists         9.5873     3.3909   2.827 0.004914 ** 
Shots                 2.2237     0.7988   2.784 0.005612 ** 
Total.Penalties     -63.3366    13.0787  -4.843 1.79e-06 ***
Minor                64.4153    13.5545   4.752 2.75e-06 ***
Takeaways             2.6666     1.7598   1.515 0.130449    
Hits.Taken           -2.3178     0.5818  -3.983 7.98e-05 ***
CF_pp                 1.8019     0.4640   3.883 0.000119 ***
CA_pp               -10.1331     5.9970  -1.690 0.091816 .  
SCA_pp               14.7489     7.3924   1.995 0.046662 *  
HDCF_pp              -5.5121     1.7483  -3.153 0.001730 ** 
LDCA_pp              19.6599     7.6634   2.565 0.010645 *  
Goals_pp            -16.2565     8.4995  -1.913 0.056464 .  
Total.Assists_pp     14.7384     5.9731   2.467 0.013998 *  
Shots_pp              9.8613     2.8874   3.415 0.000698 ***
iSCF_pp              -9.9211     3.2312  -3.070 0.002274 ** 
iHDCF_pp              9.0660     3.0494   2.973 0.003115 ** 
SCA_pk               -2.0989     1.2450  -1.686 0.092538 .  
HDCF_pk             -22.8880     7.9500  -2.879 0.004190 ** 
HDCA_pk               5.6729     3.1166   1.820 0.069429 .  
Total.Assists_pk     55.7194    26.9421   2.068 0.039231 *  
iCF_pk              -12.7809     7.6090  -1.680 0.093746 .  
iSCF_pk              19.0641    11.7231   1.626 0.104646    
Rebounds.Created_pk  96.7542    29.8999   3.236 0.001307 ** 
Penalties.Drawn_pk   15.0288     9.1585   1.641 0.101540    
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for Gamma family taken to be 0.1137775)

    Null deviance: 119.038  on 453  degrees of freedom
Residual deviance:  52.455  on 427  degrees of freedom
AIC: 14106

Number of Fisher Scoring iterations: 5

Curve Fit - Defense


Call:
glm(formula = CAP.HIT.. ~ Goals + Shots + iSCF + Hits.Taken + 
    Shots.Blocked + CF_pp + SCF_pp + LDCF_pp + Shots_pp + Rebounds.Created_pp + 
    PIM_pp + Total.Penalties_pp + Minor_pp + CA_pk + SCA_pk + 
    LDCA_pk + Total.Assists_pk + PIM_pk + Total.Penalties_pk + 
    Minor_pk + Giveaways_pk, family = Gamma(link = "log"), data = d_train_df)

Deviance Residuals: 
     Min        1Q    Median        3Q       Max  
-1.00038  -0.22529   0.01786   0.21782   0.72854  

Coefficients:
                     Estimate Std. Error t value Pr(>|t|)    
(Intercept)           15.4079     0.1647  93.533  < 2e-16 ***
Goals                -43.5053    13.6619  -3.184  0.00164 ** 
Shots                  3.7504     1.8355   2.043  0.04213 *  
iSCF                  -6.2090     2.7194  -2.283  0.02330 *  
Hits.Taken            -2.5047     0.7526  -3.328  0.00101 ** 
Shots.Blocked         -2.7418     1.1342  -2.417  0.01639 *  
CF_pp                 24.4404    10.4749   2.333  0.02047 *  
SCF_pp               -24.2866    10.8339  -2.242  0.02590 *  
LDCF_pp              -25.2514    10.9237  -2.312  0.02165 *  
Shots_pp              15.1178     5.0835   2.974  0.00324 ** 
Rebounds.Created_pp  -42.5128    18.6720  -2.277  0.02369 *  
PIM_pp                72.2512    35.0528   2.061  0.04037 *  
Total.Penalties_pp  -459.8654   191.7612  -2.398  0.01725 *  
Minor_pp             351.2200   128.4503   2.734  0.00672 ** 
CA_pk                -22.5169     7.7119  -2.920  0.00384 ** 
SCA_pk                24.6426     7.8301   3.147  0.00186 ** 
LDCA_pk               22.1086     8.2589   2.677  0.00795 ** 
Total.Assists_pk      94.3023    39.9053   2.363  0.01893 *  
PIM_pk                56.7550    25.6723   2.211  0.02801 *  
Total.Penalties_pk  -562.6875   200.4699  -2.807  0.00542 ** 
Minor_pk             356.7556   152.0071   2.347  0.01975 *  
Giveaways_pk          24.5560     8.9598   2.741  0.00660 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for Gamma family taken to be 0.1146516)

    Null deviance: 58.450  on 259  degrees of freedom
Residual deviance: 32.468  on 238  degrees of freedom
AIC: 8101

Number of Fisher Scoring iterations: 6

Curve Results - Forwards

plot of chunk unnamed-chunk-11

plot of chunk unnamed-chunk-12

Curve Results - Defense

plot of chunk unnamed-chunk-13

plot of chunk unnamed-chunk-14

Testing and Develoment Results

plot of chunk unnamed-chunk-15

Player Results - Historical - Forwards

Player	Team	year	GP	Cap_Hit	Estimated_Cap_Hit
Claude_Giroux	PHI	2016	78	8275000	9536499
Jason_Spezza	DAL	2015	82	7000000	9058724
Alex_Ovechkin	WSH	2017	82	9538462	8692597
Tyler_Seguin	DAL	2015	71	5750000	8291754
Ryan_O_Reilly	BUF	2017	72	7500000	8187712
Jason_Spezza	DAL	2016	75	7500000	8016420
Tyler_Seguin	DAL	2017	82	5750000	7953395
Phil_Kessel	TOR	2015	82	8000000	7935799
Claude_Giroux	PHI	2017	82	8275000	7836780
Alex_Ovechkin	WSH	2015	81	9538462	7798531
Phil_Kessel	PIT	2016	82	8000000	7572535
Joe_Thornton	S.J	2016	82	6750000	7527376
Jason_Spezza	DAL	2017	68	7500000	7518205
Evgeni_Malkin	PIT	2017	62	9500000	7453605

Player Results - Historical - Defense

Player	Team	year	GP	Cap_Hit	Estimated_Cap_Hit
P_K__Subban	MTL	2016	68	9000000	10954811
Keith_Yandle	FLA	2017	82	6350000	9352123
Dion_Phaneuf	TOR	2015	70	7000000	8852881
Shea_Weber	MTL	2017	78	7857143	8819301
Dougie_Hamilton	BOS	2015	72	925000	7688128
John_Carlson	WSH	2017	72	3966667	7613749
Kris_Letang	PIT	2016	71	7250000	7353238
Alexander_Edler	VAN	2015	74	5000000	7305675
Drew_Doughty	L.A	2017	82	7000000	7257779
Shea_Weber	NSH	2015	78	7857143	7082954
Aaron_Ekblad	FLA	2017	68	925000	7072747
Oliver_Ekman_Larsson	ARI	2016	75	5500000	7044832
Jake_Muzzin	L.A	2017	82	4000000	6993346
Brent_Burns	S.J	2016	82	5760000	6947536

Player Results - Current - Forward

Player	Team	year	Cap_Hit	Estimated_Cap_Hit
Jakub_Voracek	PHI	2018	8250000	17932527
Jonathan_Toews	CHI	2018	10500000	15682475
Evgeni_Malkin	PIT	2018	9500000	11020279
Evgeny_Kuznetsov	WSH	2018	7800000	9733922
Jonathan_Drouin	MTL	2018	5500000	9508431
Joe_Thornton	S.J	2018	8000000	9469489
Mike_Hoffman	OTT	2018	5187500	9319404
David_Backes	BOS	2018	6000000	9225687
Patrick_Kane	CHI	2018	10500000	9144583
Johnny_Gaudreau	CGY	2018	6750000	8738026
Patrice_Bergeron	BOS	2018	6875000	8357984
Vladimir_Tarasenko	STL	2018	7500000	8332791
Blake_Wheeler	WPG	2018	5600000	8156201
Steven_Stamkos	T.B	2018	8500000	8151298

Player Results - Current - Defense

Player	Team	year	Cap_Hit	Estimated_Cap_Hit
Dustin_Byfuglien	WPG	2018	7600000	18654542
Roman_Josi	NSH	2018	4000000	12868584
P_K__Subban	NSH	2018	9000000	12836368
Brad_Hunt	VGK	2018	650000	12238751
Duncan_Keith	CHI	2018	5538462	9839100
Sami_Vatanen	ANA	2018	4875000	9756144
Matt_Tennyson	BUF	2018	650000	9441197
Keith_Yandle	FLA	2018	6350000	9389085
Kris_Letang	PIT	2018	7250000	9269856
Shea_Weber	MTL	2018	7857143	8908570
Adam_Pelech	NYI	2018	1600000	8905920
Ben_Hutton	VAN	2018	2800000	8678523
Rasmus_Ristolainen	BUF	2018	5400000	8518337
Karl_Alzner	MTL	2018	4625000	8307787

Team Summary

plot of chunk unnamed-chunk-23

Current Team Predictions - glm


Call:
lm(formula = points ~ est_weighted_cap + LDSV. + MDSV. + HDSV., 
    data = team_summary_df)

Residuals:
    Min      1Q  Median      3Q     Max 
-33.815  -8.980   2.360   8.143  26.283 

Coefficients:
                   Estimate Std. Error t value Pr(>|t|)    
(Intercept)      -2.047e+03  4.374e+02  -4.681 1.07e-05 ***
est_weighted_cap  1.020e-05  4.095e-06   2.491  0.01470 *  
LDSV.             1.251e+01  4.417e+00   2.832  0.00578 ** 
MDSV.             6.393e+00  2.094e+00   3.053  0.00303 ** 
HDSV.             2.855e+00  9.299e-01   3.070  0.00287 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 12.15 on 85 degrees of freedom
Multiple R-squared:  0.333, Adjusted R-squared:  0.3016 
F-statistic: 10.61 on 4 and 85 DF,  p-value: 5.089e-07

Current Team Predictions

plot of chunk unnamed-chunk-25

Current Team predictions - full glm

Team	Estimated_Cap	LDSV_Perc	MDSV_Perc	HDSV_Perc	PDO	Points	Rank	Est_Rank	Predicted_Points
WSH	4579948	99.28	96.65	87.86	1.020	35	10.0	1	109.77320
CHI	5688806	98.53	95.80	88.71	1.004	29	22.0	2	108.69379
PHI	4141515	99.09	96.61	89.96	1.017	29	22.0	3	108.66507
ANA	3968388	99.52	96.87	87.55	1.013	30	19.0	4	107.05955
L.A	4044817	98.14	98.06	90.23	1.019	41	2.0	5	105.83647
S.J	4585989	99.36	96.33	86.70	0.979	32	15.5	6	105.47810
NSH	5262799	98.67	95.65	88.39	1.005	39	4.0	7	104.22788
T.B	4447205	98.94	96.11	88.68	1.029	42	1.0	8	103.05582
STL	4737125	98.67	95.99	88.32	1.013	40	3.0	9	100.84051
WPG	5372630	98.91	94.99	87.14	1.010	38	5.0	10	100.56211
VAN	4493144	97.21	97.79	90.60	1.009	32	15.5	11	98.10664
MTL	5273781	98.86	95.82	84.42	0.989	30	19.0	12	96.46904
CAR	4508526	98.82	94.62	89.79	0.991	28	24.0	13	95.82431
CBJ	3915007	98.74	96.73	87.50	1.003	35	10.0	14	95.72130

Current Team predictions - weighted salary

Team	Estimated_Cap	LDSV_Perc	MDSV_Perc	HDSV_Perc	PDO	Points	Rank	Est_Rank	Predicted_Points
CHI	5688806	98.53	95.80	88.71	1.004	29	22.0	2	108.69379
WPG	5372630	98.91	94.99	87.14	1.010	38	5.0	10	100.56211
MTL	5273781	98.86	95.82	84.42	0.989	30	19.0	12	96.46904
NSH	5262799	98.67	95.65	88.39	1.005	39	4.0	7	104.22788
CGY	4993286	97.49	97.27	87.88	0.990	32	15.5	15	95.61992
BUF	4947457	98.37	95.81	83.41	0.971	18	31.0	21	84.06481
STL	4737125	98.67	95.99	88.32	1.013	40	3.0	9	100.84051
VGK	4695134	98.06	97.19	84.47	1.004	35	10.0	18	89.46206
COL	4662615	97.71	97.23	83.64	0.989	26	26.5	23	82.63877
PIT	4611159	98.34	94.03	87.55	0.961	35	10.0	24	80.70054
S.J	4585989	99.36	96.33	86.70	0.979	32	15.5	6	105.47810
WSH	4579948	99.28	96.65	87.86	1.020	35	10.0	1	109.77320
CAR	4508526	98.82	94.62	89.79	0.991	28	24.0	13	95.82431
VAN	4493144	97.21	97.79	90.60	1.009	32	15.5	11	98.10664

Closing Notes

The full analysis is avaialable at: http://rpubs.com/john-grando/Data607_FinalProject

Be kind, rewind… and cache your web scrapes:

#Be kind, rewind.... and cache web scrapes.

load(url("https://github.com/john-grando/data607-finalproject/raw/master/html_files.RData"))