The Bahujan Samaj Party (BSP) on Monday announced that it will not continue its 2019 alliance with the Samajwadi Party (SP) in Uttar Pradesh. A series of tweets by BSP president Mayawati held the SP’s “post-poll behaviour”, which she said raised doubts on whether it can defeat the Bharatiya Janata Party (BJP), as the main factor behind the decision.She also claimed the high moral ground for the BSP, which she said, allied with the SP in “people’s interest” in spite of the latter’s poor track law-and-order track record and anti-Dalit politics during its tenure between 2012 and 2017. The BSP had earlier accused the SP of failing to ensure transfer of its core voters to BSP candidates during the 2019 elections. An HT analysis of election statistics suggests that the BSP’s decision might further strengthen the BJP in Uttar Pradesh at the cost of both the SP and the BSP, especially the latter.The BJP won 62 out of the 80 seats in Uttar Pradesh in the 2019 elections with a vote share of 49.6%. Compared to 2014, this is a decline of nine seats despite an increase of seven percentage points in vote share. A long-term analysis of BJP’s performance in Uttar Pradesh shows that the BJP’s seat share to vote share ratio in 2019 is the lowest among the four times (1996, 1998, 2014 and 2019) it has won majority of seats in the state. Seat share to vote share ratio is a useful metric to measure a party’s ability to convert popular support into seats. This analysis excludes the five parliamentary constituencies which were carved out after the creation of Uttarakhand . These numbers suggest that the SP-BSP alliance was a crucial factor in reducing the BJP’s seat tally in Uttar Pradesh despite the party getting its highest-ever vote share. An assembly constituency (AC)-wise disaggregated analysis can explain this more clearly.There were 373 ACs in Uttar Pradesh, which had BJP candidates in the 2014, 2017 and 2019 elections. The BJP won 310, 302 and 262 of these in the 2014, 2017 and 2019 elections. The trends are completely opposite for the SP and the BSP. Of the 143 ACs, which the SP contested in each of these, it won 33, 27 and 37 in the 2014, 2017 and 2019 elections. There are 192 ACs that had BSP candidates in 2014, 2017 and 2019 elections, and it won 65 of these in 2019 against just seven and 13 in 2014 and 2017. This can be seen from the comparison of strike rates – seats won as a percentage of seats contested – of the three parties. To be sure, the fact remains that the SP-BSP alliance has failed to live up to its expectations in the 2019 polls. The main reason for this is a decline in the combined vote share of the SP and BSP in 2019 compared to 2014 and 2017, even as the BJP increased its vote share to almost 50%. However, any such analysis must not ignore that the 2019 performance of the SP-BSP alliance is the best against the BJP since SP’s assembly victory in 2012 in terms of seats, even with a lower vote-share. It is also noteworthy that the SP and the BSP had performed much better in the 2017 assembly elections than the 2014 Lok Sabha elections. This suggests that an alliance could have performed better in the 2022 assembly elections in the state. Of the 192 ACs that had candidates from the BJP, the BSP and the SP in 2014 and 2017 elections and BJP versus SP-BSP candidates in 2019, the combined vote share of the SP and BSP was more than 50% in 2017 elections (See Chart 4).The reasons for poor performance of the SP-BSP alliance in 2019 are situated more in chemistry, which explains the increase in BJP’s vote share, than arithmetic. However, by calling off the alliance with the SP, the BSP may have kicked away the ladder without which it will be extremely difficult to counter the BJP in the near future.