isin, between & str.contains

Some filters come up so often that writing them with | and & gets tedious. "Players from MI, CSK or KKR" needs three == joined by |. "Runs between 200 and 450" needs two comparisons joined by &. "Names containing Sharma" can't be written with == at all. Pandas has a dedicated method for each.

Definition

Series.isin(values) returns a boolean mask that is True where the element is one of the given values. Series.between(left, right) returns True where the element lies within the range left…right (both ends included by default). Series.str.contains(pattern) returns True where a text value contains the pattern.

• All three return masks, so they plug straight into df[...] or df.loc[...] and combine with &, |, ~.
• isin is the pandas version of SQL's IN (...); between is SQL's BETWEEN ... AND ...; str.contains is like SQL's LIKE '%...%'.
• .str is the string accessor — a gateway to dozens of text methods (lower, startswith, replace, split…) that work on every value of a text column at once. It gets its own chapter later; here we use it for filtering.

Syntax

s.isin(values)                              # values: list, set, Series...
s.between(left, right, inclusive="both")    # "both", "neither", "left", "right"
s.str.contains(pat, case=True, na=None, regex=True)
s.str.startswith(prefix), s.str.endswith(suffix)

Example — isin (and NOT isin)

print(players[players["team"].isin(["MI", "CSK"])][["player", "team"]])
print(players[~players["team"].isin(["MI", "CSK"])][["player", "team"]])

            player team
1     Rohit Sharma   MI
2         MS Dhoni  CSK
3   Jasprit Bumrah   MI
4    Hardik Pandya   MI
5  Ravindra Jadeja  CSK
         player team
0   Virat Kohli  RCB
6  Shubman Gill   GT
7   Rashid Khan   GT

How it works: isin(["MI", "CSK"]) is exactly (team == "MI") | (team == "CSK"), but shorter and easy to extend to 10 teams. Putting ~ in front gives NOT IN — pandas has no notin method, so ~...isin(...) is the standard idiom.

Example — between

print(players[players["runs"].between(200, 450)][["player", "runs"]])
print(players[players["runs"].between(200, 426, inclusive="left")][["player", "runs"]])

            player  runs
1     Rohit Sharma   417
4    Hardik Pandya   216
5  Ravindra Jadeja   267
6     Shubman Gill   426
            player  runs
1     Rohit Sharma   417
4    Hardik Pandya   216
5  Ravindra Jadeja   267

How it works: between(200, 450) = (runs >= 200) & (runs <= 450) — both ends included. In the second call the upper bound is exactly Gill's 426, and inclusive="left" means "include 200 but exclude 426", so Gill drops out. between also works on dates (df["date"].between("2024-04-01", "2024-04-30")) and on strings (alphabetical range).

Example — str.contains teaser

print(players[players["player"].str.contains("Sharma")])
print(players[players["player"].str.contains("sh", case=False)][["player"]])
print(players[players["player"].str.startswith("R")][["player"]])

         player team    role  runs  wickets  strike_rate
1  Rohit Sharma   MI  Batter   417        0        150.0
         player
1  Rohit Sharma
6  Shubman Gill
7   Rashid Khan
            player
1     Rohit Sharma
5  Ravindra Jadeja
7      Rashid Khan

How it works: str.contains("Sharma") checks whether each name contains the text anywhere. With case=False the search for "sh" also matches the capital Sh in "Sharma" and "Shubman" and the "sh" in "Rashid". str.startswith("R") matches only names beginning with R. If the column has missing values, add na=False so those rows count as "no match".

Comparison — long form vs shortcut

Common mistakes

• ❌ players["team"].isin("MI") → TypeError (a single string is not a list of values). ✅ isin(["MI"]).
• ❌ Looking for a notin method. ✅ Use ~s.isin([...]).
• ❌ Forgetting that between includes both ends. ✅ Use inclusive="neither" / "left" / "right" when an end must be excluded.
• ❌ str.contains("(c)") matching unexpected rows. ✅ pat is a regular expression by default; pass regex=False for plain text with special characters like ( . + ?.
• ❌ ValueError: Cannot mask with non-boolean array containing NA / NaN values. ✅ The text column has missing values — use str.contains(..., na=False).

Interview tip

Q: How do you filter rows whose value is NOT in a list, in pandas? A: Negate isin with ~: df[~df["team"].isin(["MI", "CSK"])]. There is no notin method; the tilde inverts the boolean mask element-wise.