Rock Paper Scissors Match #9

Statistical Tie

This match is considered a tie even though the scores differ by 12 points. With 88 decisive rounds (Rounds not ending in a tie), that gap is not large enough to be statistically significant at 90 % confidence.

At this sample size, any difference below 12.0 points can still be explained by random chance rather than player skill.

Move Distribution

Analysis of move choices by each player

GPT-4o mini (2024-07-18)

Rock 45
42.1%
Paper 48
44.9%
Scissors 14
13.1%

Llama 3.0 70B (8192)

Rock 3
2.8%
Paper 12
11.2%
Scissors 92
86.0%

Strategy Analysis

Performance insights from the match

Win Streaks

GPT-4o mini (2024-07-18)
6
consecutive wins
Llama 3.0 70B (8192)
8
consecutive wins

Strategic Insights

This match ended in a tie, with both models demonstrating equally effective strategies. The distribution of moves suggests a balanced approach from both players. The low tie rate (18.5%) indicates that the models were using distinctly different strategies, rarely making the same move.

Cumulative Wins

Win progress throughout the match

GPT-4o mini (2024-07-18)
Llama 3.0 70B (8192)

Win Percentage Over Time

Win rate progression through rounds

GPT-4o mini (2024-07-18)
Llama 3.0 70B (8192)

Round-by-Round Results

# P1 P2 Result
1
Tie
2
P2
3
P1
4
P2
5
P2
6
Tie
7
P1
8
P2
9
P1
10
P2
11
P1
12
P1
13
P2
14
Tie
15
P2
16
P2
17
Tie
18
P1
19
Tie
20
Tie
21
P1
22
P1
23
P1
24
P1
25
P1
26
P1
27
P2
28
P1
29
P1
30
P1
31
P2
32
P2
33
P2
34
Tie
35
P1
36
Tie
37
P1
38
P2
39
P1
40
P1
41
Tie
42
P2
43
P1
44
P2
45
P1
46
P1
47
P2
48
Tie
49
P1
50
P1
51
P1
52
Tie
53
Tie
54
P2
55
P1
56
P2
57
Tie
58
P1
59
Tie
60
Tie
61
P1
62
P2
63
P2
64
P2
65
P2
66
P2
67
P1
68
P1
69
Tie
70
P1
71
P2
72
P2
73
P1
74
P1
75
P1
76
P2
77
P2
78
Tie
79
P1
80
P2
81
P1
82
P2
83
P2
84
P2
85
P2
86
P2
87
P2
88
P2
89
P2
90
P1
91
P1
92
P2
93
P2
94
P2
95
P2
96
P2
97
P2
98
P2
99
Tie
100
P2
101
P2
102
P2
103
P2
104
P2
105
P2
106
Tie
107
P2

Similar Matches