BUG: inserting list of strings into Series auto-infers them as datetimes with mixed formats

[MarcoGorelli]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

In [3]: df = pd.DataFrame({'a': pd.date_range('2000', freq='D', periods=2)})

In [4]: df.loc[:, 'a'] = ['12/01/2020', '13/01/2020']

In [5]: df
Out[5]:
           a
0 2020-12-01
1 2020-01-13

Issue Description

Similar to https://pandas.pydata.org/pdeps/0004-consistent-to-datetime-parsing.html

Expected Behavior

I think that inferring strings as datetimes is fine so long as they're parsed in a consistent format

Installed Versions

INSTALLED VERSIONS

commit : 57fd502
python : 3.10.12
python-bits : 64
OS : Linux
OS-release : 5.15.167.4-microsoft-standard-WSL2
Version : #1 SMP Tue Nov 5 00:21:55 UTC 2024
machine : x86_64
processor : x86_64
byteorder : little
LC_ALL : None
LANG : C.UTF-8
LOCALE : en_US.UTF-8

pandas : 3.0.0.dev0+1979.g57fd50221e
numpy : 1.26.4
dateutil : 2.9.0.post0
pip : 25.0.1
Cython : 3.0.12
sphinx : 8.1.3
IPython : 8.33.0
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : 4.13.3
blosc : None
bottleneck : 1.4.2
fastparquet : 2024.11.0
fsspec : 2025.2.0
html5lib : 1.1
hypothesis : 6.127.5
gcsfs : 2025.2.0
jinja2 : 3.1.5
lxml.etree : 5.3.1
matplotlib : 3.10.1
numba : 0.61.0
numexpr : 2.10.2
odfpy : None
openpyxl : 3.1.5
psycopg2 : 2.9.10
pymysql : 1.4.6
pyarrow : 19.0.1
pyreadstat : 1.2.8
pytest : 8.3.5
python-calamine : None
pytz : 2025.1
pyxlsb : 1.0.10
s3fs : 2025.2.0
scipy : 1.15.2
sqlalchemy : 2.0.38
tables : 3.10.1
tabulate : 0.9.0
xarray : 2024.9.0
xlrd : 2.0.1
xlsxwriter : 3.2.2
zstandard : 0.23.0
tzdata : 2025.1
qtpy : None
pyqt5 : None

[MarcoGorelli]

sorry how's that related?

[Farsidetfs]

It's not just with loc.
same results:

df = DataFrame({'a': ['12/01/2020', '13/01/2020']}, dtype='datetime64[ns]')

0 2020-12-01
1 2020-01-13

[iabhi4]

take

[iabhi4]

@MarcoGorelli Hey, I looked into this and the root issue seems to be that ambiguous datetime strings like "12/01/2020" and "13/01/2020" are being silently parsed inconsistently — one as MM/DD/YYYY and the other as DD/MM/YYYY — depending on what's valid. This happens both when assigning with .loc and during DataFrame construction with dtype='datetime64[ns]', since NumPy ends up handling the coercion without format checks.

I added a small check inside maybe_coerce_values() that kicks in when we’re dealing with a 1D object array of strings. It tries to infer a consistent datetime format, and raises a ValueError if the format is ambiguous or inconsistent. This is based on the idea mentioned in the issue that inferring is okay as long as the format is consistent.

Would this kind of fix be reasonable? If so, I can open a PR. Let me know!

[MarcoGorelli]

yeah that might be fine

[iabhi4]

yeah that might be fine

just opened a PR. When you have a moment, would you mind taking a look? Appreciate your input!

[MarcoGorelli]

i don't really have much capacity for reviews, but i'll note that your solution probably adds more complexity than it warranted, if you want it to be merged it should be simpler and cleaner