What last name has appeared in the MLB each year for the longest time?

baseball_last_names.py

from collections import defaultdict

import pandas as pd


# Data is from seanlahman.com
people = pd.read_csv(
    "People.csv", 
    encoding="latin-1",
    low_memory=False
)

appearances = pd.read_csv(
    "Appearances.csv",
    encoding="latin-1",
    low_memory=False
)

player_id_to_last_name: dict[str, str] = {}
last_name_to_ids: defaultdict[str, set[str]] = defaultdict(set)

for _, row in people.iterrows():
    pid = row["playerID"]
    last_name = row["nameLast"]
    player_id_to_last_name[pid] = last_name
    last_name_to_ids[last_name].add(pid)

player_id_to_years: defaultdict[str, set[int]] = defaultdict(set)
last_name_to_years: defaultdict[str, set[int]] = defaultdict(set)

for _, row in appearances.iterrows():
    pid = row["playerID"]
    year = int(row["yearID"])
    player_id_to_years[pid].add(year)
    last_name = player_id_to_last_name.get(pid)
    last_name_to_years[last_name].add(year)


possible_last_names = set(last_name_to_ids)
last_name_to_start: dict[str, int] = {}

for start_year in range(1871, 2024):
    target_years = set(range(start_year, 2024))
    # the names where all the target years have someone with that last name
    last_names_full_coverage: set[str] = set()
    for last_name in list(possible_last_names):
        years = last_name_to_years[last_name]
        if years.issuperset(target_years):
            last_names_full_coverage.add(last_name)
            possible_last_names.remove(last_name)
            last_name_to_start[last_name] = start_year
print("2023")
top = sorted(last_name_to_start.items(), key=lambda x: x[1])[:10]
for i, (name, year) in enumerate(top, start=1):
    print(f"{i}. {name}: {year}")
    
"""
2023
1. Johnson: 1907
2. Miller: 1939
3. Jones: 1945
4. Jackson: 1950
5. Smith: 1950
6. Davis: 1951
7. Reed: 1958
8. Martinez: 1962
9. Perez: 1964
10. Hernandez: 1965
"""

# Redo it and remove Reed, since that was a top 10 but there was no Reed in 2024

possible_last_names = set(last_name_to_ids) - {"Reed"}
last_name_to_start: dict[str, int] = {}

for start_year in range(1871, 2024):
    target_years = set(range(start_year, 2024))
    # the names where all the target years have someone with that last name
    last_names_full_coverage: set[str] = set()
    for last_name in list(possible_last_names):
        years = last_name_to_years[last_name]
        if years.issuperset(target_years):
            last_names_full_coverage.add(last_name)
            possible_last_names.remove(last_name)
            last_name_to_start[last_name] = start_year
print("\n2024")
top = sorted(last_name_to_start.items(), key=lambda x: x[1])[:10]
for i, (name, year) in enumerate(top, start=1):
    print(f"{i}. {name}: {year}")
    
"""
2024
1. Johnson: 1907
2. Miller: 1939
3. Jones: 1945
4. Jackson: 1950
5. Smith: 1950
6. Davis: 1951
7. Martinez: 1962
8. Perez: 1964
9. Hernandez: 1965
10. Rodriguez: 1967
"""

# code gets less good after here
in_streak_last_name_to_ids: dict[str, set[str]] = defaultdict(set)

for name, start_year in last_name_to_start.items():
    streak = set(range(start_year, 2024))
    for pid in last_name_to_ids[name]:
        played_years = player_id_to_years[pid]
        if not played_years.isdisjoint(streak):
            in_streak_last_name_to_ids[name].add(pid)
            
long_streaks_few_players = []
for name, start_year in last_name_to_start.items():
    player_count = len(in_streak_last_name_to_ids[name])
    long_streaks_few_players.append((player_count, start_year, name))
long_streaks_few_players.sort()

count_to_year = {}
for player_count, start_year, name in long_streaks_few_players:
    if player_count not in count_to_year:
        count_to_year[player_count] = start_year
        
keepers = []
x =  sorted([(count, year) for count, year in count_to_year.items()])
for count, year in x:
    if not keepers or year < keepers[-1][1]:
        keepers.append((count, year))
count_to_year_keepers = dict(keepers)
    
count_year_name = defaultdict(list)    
for player_count, start_year, name in long_streaks_few_players:
    if count_to_year_keepers.get(player_count) == start_year:
        count_year_name[(player_count, start_year)].append(name)
        
for (count, start_year), names in count_year_name.items():
    names = ', '.join(sorted(names))
    print(f"{start_year}: {count} {names}")

"""
2004: 1 Greinke
1998: 3 Polanco
1997: 4 Gomes
1996: 6 Chavez
1991: 7 Hunter
1985: 9 Myers
1978: 13 Stewart
1973: 17 Hill
1967: 20 Harris
1950: 33 Jackson
1939: 51 Miller
1907: 108 Johnson
"""