Karnataka Elections and role of BSP

So the election results for Karnataka assembly were kind of as predicted as well as surprising. BJP finally got almost a majority and they probably would peacefully run a government for next five years. But most of the analysts on TV were (without looking at the data) saying that congress lost because of BSP. So I thought that let me see if the data supports that fact. Here is a table of constituencies where congress's loss margin was less than the vote that BSP got. For simplicity I am assuming that if BSP was not the spoilsport and all the vote would have gone to congress they would have won. Here is the data

S No.	Constituency No	Constituency Name	Votes to Congress	Votes polled to BSP	Votes of Winner	Victory Margin
1	32	Indi	28885	7840	29456	571
2	35	Jevargi	46461	2266	46531	70
3	46	Aland	36689	14405	42473	5784
4	71	Dharwad	34694	1302	35417	723
5	78	Kumta	30772	1819	30792	20
6	81	Yellapur	36624	3278	39109	2485
7	98	Challakere	42302	3938	42591	289
8	109	Channagiri	38533	1372	39526	993
9	121	Kapu	44994	1109	45961	967
10	123	Sringeri	41396	4758	43646	2250
11	124	Mudigere	26084	11724	34579	8495
12	129	Tiptur	39168	8021	46034	6866
13	138	Madhugiri	51408	769	51971	563
14	153	Yeshvanthapura	56561	1331	57643	1082
15	206	Puttur	45180	2553	46605	1425

So there seems to be some merit to the logic that there is a significant number of seats that could have gone the congress way if all the vote that BSP got would have got transferred to congress. But this is only half the story. So I collected the same data for BJP and here are the results.

S No.	Constituency No	Constituency Name	Votes to BJP	Votes polled to BSP	Votes of Winner	Victory Margin
1	5	Kudachi	28715	1910	29481	766
2	18	Ramdurg	48862	1187	49246	384
3	26	Muddebihal	21662	14286	24065	2403
4	49	Bidar South	30783	21907	32054	1271
5	50	Bidar	20561	30627	33557	12996
6	53	Raichur Rural	29097	8499	34432	5335
7	75	Kalghatgi	38091	14992	49733	11642
8	78	Kumta	30201	1819	30792	591
9	103	Jagalur	35873	6482	38664	2791
10	205	Bantval	60309	1396	61560	1251
11	214	Nanjangud	42159	4513	42867	708
12	222	Kollegal	25586	25505	37384	11798
13	223	Chamarajanagar	39405	8866	42017	2612
14	224	Gundlupet	62621	7677	64824	2203

And the result and the damage is very similar, Now one can not really make a call on whom does BSP vote go to if the BSP was not contesting the election and I would think that it is more likely to get transferred in the ratio of the votes that were otherwise polled to each of the parties. A similar analysis for JDS also shows following

S No.	Constituency No	Constituency Name	Votes to BJP	Votes polled to BSP	Votes of Winner	Victory Margin
1	26	Muddebihal	14739	14286	24065	9326
2	53	Raichur Rural	32555	8499	34432	1877
3	112	Bhadravati	52770	2532	53257	487
4	127	Kadur	36000	4078	39411	3411
5	143	Chintamani	56857	1419	58103	1246
6	192	Krishnarajpet	45500	10218	48556	3056
7	210	Piriyapatna	37574	7814	38453	879
8	222	Kollegal	25286	25505	37384	12098

Another eight such constituencies. Now in each of these datasets the party in question (i.e. Congress, BJP, BSP) is not necessarily the second party, in some cases it may be third.

One thing is certain, BSP got significant percentage of votes in certain seats and definitely impacted the result of election but I would not really believe that it has impacted one party more than the other.