NFL offense performance and analysis#

Author: Shengkai Yang

Course Project, UC Irvine, Math 10, F22

Introduction#

In this project, I want to use machine learning to analyze how some specific data which affect NFL football games. Also, I want to know how these data affect teams performance and predict them.

Main part of project#

You can either have all one section or divide into multiple sections. To make new sections, use ## in a markdown cell. Double-click this cell for an example of using ##

import pandas as pd
import numpy as np
import altair as alt
import matplotlib.pyplot as plt
from sklearn.cluster import KMeans
import seaborn as sns

w4 = pd.read_csv("nfloffenseweek4.csv")

w4

	rank	team	games	points_scored	total_yards	offensive_plays	yards_per_play	turnovers_lost	fumbles_lost	1st_downs	...	rushing_yards	rushing_touchdowns	rushing_yards_per_attempt	rushing_1st_downs	penalties	penalty_yards	1st_down_penalties	percentage_scoring_drives	percentage_turnover_drives	expected_points
0	1	Detroit Lions	4	140	1747	269	6.5	4	1	90	...	656	7	5.9	26	23	188	7	45.8	8.3	53.17
1	2	Kansas City Chiefs	4	129	1539	257	6.0	4	2	95	...	468	4	4.5	23	19	156	10	50.0	7.1	57.85
2	3	Baltimore Ravens	4	119	1437	230	6.2	5	1	77	...	568	3	5.4	31	17	114	7	42.2	11.1	32.97
3	4	Philadelphia Eagles	4	115	1742	285	6.1	2	0	98	...	661	10	4.3	40	26	193	9	40.0	4.4	44.88
4	5	Buffalo Bills	4	114	1650	275	6.0	7	4	99	...	462	2	4.8	29	24	167	7	47.5	17.5	49.56
5	6	Cleveland Browns	4	105	1539	281	5.5	3	1	96	...	749	7	5.0	46	26	184	7	46.3	7.3	40.78
6	7	Jacksonville Jaguars	4	105	1346	250	5.4	6	4	83	...	441	3	4.0	23	22	157	8	40.9	13.6	17.69
7	8	Atlanta Falcons	4	103	1396	236	5.9	8	4	86	...	672	6	5.1	36	16	114	12	44.2	18.6	30.77
8	9	Miami Dolphins	4	98	1444	227	6.4	4	0	80	...	277	2	3.5	15	22	132	9	42.1	10.5	46.01
9	10	Las Vegas Raiders	4	96	1425	256	5.6	5	1	87	...	452	2	5.0	26	23	148	5	50.0	12.5	23.26
10	11	Seattle Seahawks	4	95	1444	228	6.3	6	3	83	...	459	3	5.2	26	32	306	9	44.1	11.8	42.94
11	12	Los Angeles Chargers	4	92	1487	264	5.6	4	2	77	...	258	2	2.7	17	21	161	5	36.4	6.8	17.02
12	13	Cincinnati Bengals	4	91	1387	290	4.8	6	2	87	...	358	1	3.1	18	22	166	11	37.5	12.5	6.37
13	14	Arizona Cardinals	4	88	1398	292	4.8	2	0	89	...	448	4	4.1	33	30	280	6	36.6	4.9	8.95
14	15	Minnesota Vikings	4	86	1376	254	5.4	5	1	92	...	392	3	4.4	22	16	95	11	38.6	11.4	20.12
15	16	Tampa Bay Buccaneers	4	82	1268	245	5.2	6	5	77	...	261	1	3.1	13	24	206	9	34.8	10.9	-3.64
16	17	Carolina Panthers	4	78	1049	214	4.9	6	3	56	...	385	2	4.5	22	23	250	7	31.1	11.1	-25.26
17	18	New Orleans Saints	4	76	1457	244	6.0	11	6	74	...	446	4	5.1	23	34	319	5	27.1	20.8	-11.00
18	19	New York Giants	4	76	1328	256	5.2	5	2	80	...	770	4	5.7	43	31	219	9	35.6	8.9	6.34
19	20	New York Jets	4	76	1458	289	5.0	9	4	82	...	350	1	4.1	15	25	257	8	32.6	19.6	-10.79
20	21	Green Bay Packers	4	75	1510	259	5.8	7	4	85	...	580	3	5.0	34	17	140	5	28.3	15.2	16.07
21	22	Tennessee Titans	4	75	1150	220	5.2	6	3	69	...	409	4	3.8	19	29	262	9	31.0	9.5	2.39
22	23	Pittsburgh Steelers	4	74	1115	234	4.8	7	2	69	...	389	4	4.0	25	24	159	4	30.4	13.0	-15.59
23	24	New England Patriots	4	74	1365	241	5.7	9	4	74	...	514	5	4.5	33	20	161	3	28.6	21.4	-0.64
24	25	Houston Texans	4	73	1208	237	5.1	5	1	68	...	380	2	4.5	19	26	226	11	31.3	10.4	-12.95
25	26	Washington Commanders	4	73	1323	287	4.6	7	2	84	...	402	2	4.1	24	23	264	8	21.6	13.7	-16.38
26	27	Dallas Cowboys	4	71	1251	241	5.2	2	1	69	...	416	2	4.0	22	27	208	7	35.6	4.4	1.65
27	28	San Francisco 49ers	4	71	1298	238	5.5	5	3	68	...	541	4	4.4	29	24	190	8	27.3	11.4	-1.21
28	29	Los Angeles Rams	4	70	1176	248	4.7	9	3	79	...	274	3	3.3	17	14	131	6	35.9	23.1	-8.08
29	30	Denver Broncos	4	66	1343	247	5.4	4	3	66	...	438	2	4.2	20	37	286	7	31.8	9.1	-6.92
30	31	Chicago Bears	4	64	1099	219	5.0	7	3	59	...	709	4	5.2	34	21	171	8	31.8	13.6	-5.87
31	32	Indianapolis Colts	4	57	1359	270	5.0	9	4	82	...	351	1	3.5	22	23	203	5	25.0	20.5	-18.74

32 rows × 28 columns

First of all, I need to clean the data, based on the original data, I have the data of rushing touchdowns and passing touchdowns, I want to collect all of them together to make sure how many touchdowns they made.

w4["touchdowns"] = w4["passing_touchdowns"]+w4["rushing_touchdowns"]
w4

	rank	team	games	points_scored	total_yards	offensive_plays	yards_per_play	turnovers_lost	fumbles_lost	1st_downs	...	rushing_touchdowns	rushing_yards_per_attempt	rushing_1st_downs	penalties	penalty_yards	1st_down_penalties	percentage_scoring_drives	percentage_turnover_drives	expected_points	touchdowns
0	1	Detroit Lions	4	140	1747	269	6.5	4	1	90	...	7	5.9	26	23	188	7	45.8	8.3	53.17	18
1	2	Kansas City Chiefs	4	129	1539	257	6.0	4	2	95	...	4	4.5	23	19	156	10	50.0	7.1	57.85	15
2	3	Baltimore Ravens	4	119	1437	230	6.2	5	1	77	...	3	5.4	31	17	114	7	42.2	11.1	32.97	14
3	4	Philadelphia Eagles	4	115	1742	285	6.1	2	0	98	...	10	4.3	40	26	193	9	40.0	4.4	44.88	14
4	5	Buffalo Bills	4	114	1650	275	6.0	7	4	99	...	2	4.8	29	24	167	7	47.5	17.5	49.56	12
5	6	Cleveland Browns	4	105	1539	281	5.5	3	1	96	...	7	5.0	46	26	184	7	46.3	7.3	40.78	11
6	7	Jacksonville Jaguars	4	105	1346	250	5.4	6	4	83	...	3	4.0	23	22	157	8	40.9	13.6	17.69	11
7	8	Atlanta Falcons	4	103	1396	236	5.9	8	4	86	...	6	5.1	36	16	114	12	44.2	18.6	30.77	9
8	9	Miami Dolphins	4	98	1444	227	6.4	4	0	80	...	2	3.5	15	22	132	9	42.1	10.5	46.01	11
9	10	Las Vegas Raiders	4	96	1425	256	5.6	5	1	87	...	2	5.0	26	23	148	5	50.0	12.5	23.26	8
10	11	Seattle Seahawks	4	95	1444	228	6.3	6	3	83	...	3	5.2	26	32	306	9	44.1	11.8	42.94	9
11	12	Los Angeles Chargers	4	92	1487	264	5.6	4	2	77	...	2	2.7	17	21	161	5	36.4	6.8	17.02	11
12	13	Cincinnati Bengals	4	91	1387	290	4.8	6	2	87	...	1	3.1	18	22	166	11	37.5	12.5	6.37	9
13	14	Arizona Cardinals	4	88	1398	292	4.8	2	0	89	...	4	4.1	33	30	280	6	36.6	4.9	8.95	9
14	15	Minnesota Vikings	4	86	1376	254	5.4	5	1	92	...	3	4.4	22	16	95	11	38.6	11.4	20.12	9
15	16	Tampa Bay Buccaneers	4	82	1268	245	5.2	6	5	77	...	1	3.1	13	24	206	9	34.8	10.9	-3.64	7
16	17	Carolina Panthers	4	78	1049	214	4.9	6	3	56	...	2	4.5	22	23	250	7	31.1	11.1	-25.26	6
17	18	New Orleans Saints	4	76	1457	244	6.0	11	6	74	...	4	5.1	23	34	319	5	27.1	20.8	-11.00	9
18	19	New York Giants	4	76	1328	256	5.2	5	2	80	...	4	5.7	43	31	219	9	35.6	8.9	6.34	7
19	20	New York Jets	4	76	1458	289	5.0	9	4	82	...	1	4.1	15	25	257	8	32.6	19.6	-10.79	8
20	21	Green Bay Packers	4	75	1510	259	5.8	7	4	85	...	3	5.0	34	17	140	5	28.3	15.2	16.07	9
21	22	Tennessee Titans	4	75	1150	220	5.2	6	3	69	...	4	3.8	19	29	262	9	31.0	9.5	2.39	9
22	23	Pittsburgh Steelers	4	74	1115	234	4.8	7	2	69	...	4	4.0	25	24	159	4	30.4	13.0	-15.59	6
23	24	New England Patriots	4	74	1365	241	5.7	9	4	74	...	5	4.5	33	20	161	3	28.6	21.4	-0.64	8
24	25	Houston Texans	4	73	1208	237	5.1	5	1	68	...	2	4.5	19	26	226	11	31.3	10.4	-12.95	7
25	26	Washington Commanders	4	73	1323	287	4.6	7	2	84	...	2	4.1	24	23	264	8	21.6	13.7	-16.38	10
26	27	Dallas Cowboys	4	71	1251	241	5.2	2	1	69	...	2	4.0	22	27	208	7	35.6	4.4	1.65	6
27	28	San Francisco 49ers	4	71	1298	238	5.5	5	3	68	...	4	4.4	29	24	190	8	27.3	11.4	-1.21	7
28	29	Los Angeles Rams	4	70	1176	248	4.7	9	3	79	...	3	3.3	17	14	131	6	35.9	23.1	-8.08	7
29	30	Denver Broncos	4	66	1343	247	5.4	4	3	66	...	2	4.2	20	37	286	7	31.8	9.1	-6.92	6
30	31	Chicago Bears	4	64	1099	219	5.0	7	3	59	...	4	5.2	34	21	171	8	31.8	13.6	-5.87	6
31	32	Indianapolis Colts	4	57	1359	270	5.0	9	4	82	...	1	3.5	22	23	203	5	25.0	20.5	-18.74	6

32 rows × 29 columns

w4["lost"] = w4["turnovers_lost"]+w4["fumbles_lost"]
w4

	rank	team	games	points_scored	total_yards	offensive_plays	yards_per_play	turnovers_lost	fumbles_lost	1st_downs	...	rushing_yards_per_attempt	rushing_1st_downs	penalties	penalty_yards	1st_down_penalties	percentage_scoring_drives	percentage_turnover_drives	expected_points	touchdowns	lost
0	1	Detroit Lions	4	140	1747	269	6.5	4	1	90	...	5.9	26	23	188	7	45.8	8.3	53.17	18	5
1	2	Kansas City Chiefs	4	129	1539	257	6.0	4	2	95	...	4.5	23	19	156	10	50.0	7.1	57.85	15	6
2	3	Baltimore Ravens	4	119	1437	230	6.2	5	1	77	...	5.4	31	17	114	7	42.2	11.1	32.97	14	6
3	4	Philadelphia Eagles	4	115	1742	285	6.1	2	0	98	...	4.3	40	26	193	9	40.0	4.4	44.88	14	2
4	5	Buffalo Bills	4	114	1650	275	6.0	7	4	99	...	4.8	29	24	167	7	47.5	17.5	49.56	12	11
5	6	Cleveland Browns	4	105	1539	281	5.5	3	1	96	...	5.0	46	26	184	7	46.3	7.3	40.78	11	4
6	7	Jacksonville Jaguars	4	105	1346	250	5.4	6	4	83	...	4.0	23	22	157	8	40.9	13.6	17.69	11	10
7	8	Atlanta Falcons	4	103	1396	236	5.9	8	4	86	...	5.1	36	16	114	12	44.2	18.6	30.77	9	12
8	9	Miami Dolphins	4	98	1444	227	6.4	4	0	80	...	3.5	15	22	132	9	42.1	10.5	46.01	11	4
9	10	Las Vegas Raiders	4	96	1425	256	5.6	5	1	87	...	5.0	26	23	148	5	50.0	12.5	23.26	8	6
10	11	Seattle Seahawks	4	95	1444	228	6.3	6	3	83	...	5.2	26	32	306	9	44.1	11.8	42.94	9	9
11	12	Los Angeles Chargers	4	92	1487	264	5.6	4	2	77	...	2.7	17	21	161	5	36.4	6.8	17.02	11	6
12	13	Cincinnati Bengals	4	91	1387	290	4.8	6	2	87	...	3.1	18	22	166	11	37.5	12.5	6.37	9	8
13	14	Arizona Cardinals	4	88	1398	292	4.8	2	0	89	...	4.1	33	30	280	6	36.6	4.9	8.95	9	2
14	15	Minnesota Vikings	4	86	1376	254	5.4	5	1	92	...	4.4	22	16	95	11	38.6	11.4	20.12	9	6
15	16	Tampa Bay Buccaneers	4	82	1268	245	5.2	6	5	77	...	3.1	13	24	206	9	34.8	10.9	-3.64	7	11
16	17	Carolina Panthers	4	78	1049	214	4.9	6	3	56	...	4.5	22	23	250	7	31.1	11.1	-25.26	6	9
17	18	New Orleans Saints	4	76	1457	244	6.0	11	6	74	...	5.1	23	34	319	5	27.1	20.8	-11.00	9	17
18	19	New York Giants	4	76	1328	256	5.2	5	2	80	...	5.7	43	31	219	9	35.6	8.9	6.34	7	7
19	20	New York Jets	4	76	1458	289	5.0	9	4	82	...	4.1	15	25	257	8	32.6	19.6	-10.79	8	13
20	21	Green Bay Packers	4	75	1510	259	5.8	7	4	85	...	5.0	34	17	140	5	28.3	15.2	16.07	9	11
21	22	Tennessee Titans	4	75	1150	220	5.2	6	3	69	...	3.8	19	29	262	9	31.0	9.5	2.39	9	9
22	23	Pittsburgh Steelers	4	74	1115	234	4.8	7	2	69	...	4.0	25	24	159	4	30.4	13.0	-15.59	6	9
23	24	New England Patriots	4	74	1365	241	5.7	9	4	74	...	4.5	33	20	161	3	28.6	21.4	-0.64	8	13
24	25	Houston Texans	4	73	1208	237	5.1	5	1	68	...	4.5	19	26	226	11	31.3	10.4	-12.95	7	6
25	26	Washington Commanders	4	73	1323	287	4.6	7	2	84	...	4.1	24	23	264	8	21.6	13.7	-16.38	10	9
26	27	Dallas Cowboys	4	71	1251	241	5.2	2	1	69	...	4.0	22	27	208	7	35.6	4.4	1.65	6	3
27	28	San Francisco 49ers	4	71	1298	238	5.5	5	3	68	...	4.4	29	24	190	8	27.3	11.4	-1.21	7	8
28	29	Los Angeles Rams	4	70	1176	248	4.7	9	3	79	...	3.3	17	14	131	6	35.9	23.1	-8.08	7	12
29	30	Denver Broncos	4	66	1343	247	5.4	4	3	66	...	4.2	20	37	286	7	31.8	9.1	-6.92	6	7
30	31	Chicago Bears	4	64	1099	219	5.0	7	3	59	...	5.2	34	21	171	8	31.8	13.6	-5.87	6	10
31	32	Indianapolis Colts	4	57	1359	270	5.0	9	4	82	...	3.5	22	23	203	5	25.0	20.5	-18.74	6	13

32 rows × 30 columns

Using Altair chart to show the relationship between their offense performance and the rank#

c1 = alt.Chart(w4).mark_circle().encode(
    x="yards_per_play",
    y="points_scored",
    color =alt.Color("rank", scale=alt.Scale(scheme="goldgreen")),
    tooltip =["team", "yards_per_play", "points_scored"]
)

c1

w4.columns

Index(['rank', 'team', 'games', 'points_scored', 'total_yards',
       'offensive_plays', 'yards_per_play', 'turnovers_lost', 'fumbles_lost',
       '1st_downs', 'passes_completed', 'passes_attempted', 'passing_yards',
       'passing_touchdowns', 'passing_interceptions',
       'net_yards_per_pass_attempt', 'passing_1st_downs', 'rushing_attempts',
       'rushing_yards', 'rushing_touchdowns', 'rushing_yards_per_attempt',
       'rushing_1st_downs', 'penalties', 'penalty_yards', '1st_down_penalties',
       'percentage_scoring_drives', 'percentage_turnover_drives',
       'expected_points', 'touchdowns', 'lost'],
      dtype='object')

from above altair chart, we can tell lighter yellow means higher rank, darker green means lower rank, in the middle part. Since Green and Yello is totally different, so we can directly see the relationship between rank and offense performance. Miami Dolphins gets top yards_per_play in the league, but the rank is low.

Using DecisionTree to classify to predict#

I want to find the relationship between yards per play and teams’ touchdowns, so I use decision tree to predict them.

from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import train_test_split
from sklearn.tree import DecisionTreeClassifier
from pandas.api.types import is_numeric_dtype

w4_1 = w4[["points_scored","yards_per_play","offensive_plays","passes_completed","passes_attempted","touchdowns"]]
w4_1

	points_scored	yards_per_play	offensive_plays	passes_completed	passes_attempted	touchdowns
0	140	6.5	269	93	152	18
1	129	6.0	257	97	147	15
2	119	6.2	230	76	117	14
3	115	6.1	285	82	123	14
4	114	6.0	275	113	170	12
5	105	5.5	281	82	127	11
6	105	5.4	250	88	134	11
7	103	5.9	236	57	98	9
8	98	6.4	227	94	140	11
9	96	5.6	256	95	155	8
10	95	6.3	228	102	133	9
11	92	5.6	264	111	166	11
12	91	4.8	290	101	157	9
13	88	4.8	292	115	177	9
14	86	5.4	254	100	158	9
15	82	5.2	245	106	155	7
16	78	4.9	214	64	117	6
17	76	6.0	244	93	143	9
18	76	5.2	256	68	108	7
19	76	5.0	289	111	193	8
20	75	5.8	259	93	134	9
21	75	5.2	220	68	105	9
22	74	4.8	234	79	129	6
23	74	5.7	241	79	118	8
24	73	5.1	237	88	142	7
25	73	4.6	287	107	172	10
26	71	5.2	241	76	131	6
27	71	5.5	238	62	108	7
28	70	4.7	248	106	150	7
29	66	5.4	247	80	131	6
30	64	5.0	219	34	67	6
31	57	5.0	270	102	154	6

num_cols = [c for c in w4_1.columns if is_numeric_dtype(w4[c])]
num_cols

['points_scored',
 'yards_per_play',
 'offensive_plays',
 'passes_completed',
 'passes_attempted',
 'touchdowns']

features = [c for c in w4_1 if c != "touchdowns"]
features

['points_scored',
 'yards_per_play',
 'offensive_plays',
 'passes_completed',
 'passes_attempted']

x = w4_1[features]
y = w4_1["touchdowns"]

X_train, X_test, y_train, y_test = train_test_split(w4_1, w4_1["touchdowns"], test_size=0.2, random_state=0)

clf = DecisionTreeClassifier(max_depth=6)

X_test
y_test

  11
   6
   9
   14
   6
   9
   7
Name: touchdowns, dtype: int64

X_train
y_train

   6
   9
   9
   7
   11
   9
   11
   6
  10
   8
   15
   6
   11
   12
   7
   6
   8
    8
    9
   7
   14
   18
   9
   7
   9
Name: touchdowns, dtype: int64

clf.fit(X_train, y_train)

DecisionTreeClassifier(max_depth=6)

clf.score(X_train, y_train)

0.96

clf.score(X_test, y_test)

0.8571428571428571

from sklearn.tree import plot_tree
import matplotlib.pyplot as plt

fig = plt.figure(figsize=(9,15))
plot_tree(
        clf,
        feature_names=clf.feature_names_in_,
        filled=True
        
    );

Based on above information, I choose max_depth is 6, and the train set value is 0.96 and the test value is round to 0.86, so this is not overfitting in this model. Also in the above Decission chart, we can see the different situation’s result. Like touchdowns less or equal than 9.5,etc.

Using K-Neighbors to predict#

This is the extra part of Math10, because this classifier is supervised and make classificartions, predictions about individual data point in a group, so I think it is good for analyze NFL datas. Even though peopla call it “lazy”.

from sklearn.neighbors import KNeighborsClassifier

scaler = StandardScaler()
scaler.fit(X)
X_scaled = scaler.transform(X)

X = w4[["rushing_attempts","passes_attempted"]]
y = w4["points_scored"]

X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, train_size=0.6, random_state=4)

clf2 = KNeighborsClassifier()
clf2.fit(X_train, y_train)

KNeighborsClassifier()

w4_1["pred"] = clf2.predict(X_scaled)
clf2.fit(X_train, y_train)

/shared-libs/python3.7/py-core/lib/python3.7/site-packages/ipykernel_launcher.py:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  """Entry point for launching an IPython kernel.

KNeighborsClassifier()

w4_1

	points_scored	yards_per_play	offensive_plays	passes_completed	passes_attempted	touchdowns	pred_1	predict	pred
0	140	6.5	269	93	152	18	57	57	57
1	129	6.0	257	97	147	15	57	57	57
2	119	6.2	230	76	117	14	66	66	66
3	115	6.1	285	82	123	14	105	105	105
4	114	6.0	275	113	170	12	57	57	57
5	105	5.5	281	82	127	11	74	74	74
6	105	5.4	250	88	134	11	66	66	66
7	103	5.9	236	57	98	9	64	64	64
8	98	6.4	227	94	140	11	73	73	73
9	96	5.6	256	95	155	8	57	57	57
10	95	6.3	228	102	133	9	66	66	66
11	92	5.6	264	111	166	11	57	57	57
12	91	4.8	290	101	157	9	57	57	57
13	88	4.8	292	115	177	9	57	57	57
14	86	5.4	254	100	158	9	57	57	57
15	82	5.2	245	106	155	7	73	73	73
16	78	4.9	214	64	117	6	66	66	66
17	76	6.0	244	93	143	9	73	73	73
18	76	5.2	256	68	108	7	64	64	64
19	76	5.0	289	111	193	8	73	73	73
20	75	5.8	259	93	134	9	66	66	66
21	75	5.2	220	68	105	9	66	66	66
22	74	4.8	234	79	129	6	66	66	66
23	74	5.7	241	79	118	8	66	66	66
24	73	5.1	237	88	142	7	73	73	73
25	73	4.6	287	107	172	10	57	57	57
26	71	5.2	241	76	131	6	66	66	66
27	71	5.5	238	62	108	7	66	66	66
28	70	4.7	248	106	150	7	73	73	73
29	66	5.4	247	80	131	6	66	66	66
30	64	5.0	219	34	67	6	64	64	64
31	57	5.0	270	102	154	6	57	57	57

c3 = alt.Chart(w4_1).mark_circle().encode(
    x="passes_completed",
    y="passes_attempted",
    color=alt.Color("pred", title="rank"),
    tooltip = ('passes_attempted','touchdowns','points_scored','yards_per_play','passes_completed')
).properties(
    title="Passing",
    width=400,
    height=400,
)
c3

Using KNeighborsCalssifier and altair chart to show the performance of passing in each team, except two teams, which is almost 50% passing complete on the left side, and another one 193 passes attempts and 111 completed, high passes, high rate of success. other team has the similar rate of passing success. We can see there is a line in the graph.

Summary#

In the final project, I use altair chart, Decission Tree and KNeighbor classifier to analyze the NFL teams offense performance. Also, using Decission Tree to make sure whtether it is overfitting or not which is very important in machine learning.

References#

Your code above should include references. Here is some additional space for references.

What is the source of your dataset(s)? https://www.kaggle.com/datasets/kendallgillies/nflstatistics

List any other references that you found helpful. https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html

Submission#

Using the Share button at the top right, enable Comment privileges for anyone with a link to the project. Then submit that link on Canvas.

Created in Deepnote

UC Irvine Math 10, Fall 2022

NFL offense performance and analysis

Contents